T

Previous page
Table of content
Next page

MAC address

A Layer 2 address for a network node.

Overview

On an Ethernet network, a MAC address is a unique 6-byte (48-bit) address that identifies a computer, router interface, or other node on the network. The MAC in the term MAC address stands for media access control, a sublayer of the Layer 2 or data-link layer of the Open Systems Interconnection (OSI) reference model for networking. MAC addresses are sometimes called Ethernet addresses, hardware addresses, or physical addresses.

MAC addresses enable devices on a network to communicate with each other. On an Ethernet network, for example, each frame contains a source MAC address and destination MAC address in its header information. Devices such as bridges and Layer 2 switches examine the source MAC address of every frame received and then use these MAC addresses to dynamically build internal routing tables for directing further traffic. They then use these tables to forward future frames received to the specific port or network segment on which the station having that particular destination MAC address resides.

Implementation

On a computer, the MAC address is assigned to the network interface card (NIC) that is used to connect the computer to the network. The MAC address is typically hard-coded into the card's read-only memory (ROM). Some cards come with a software utility that you can use to change the card's MAC address, but this is generally not a good idea. If you accidentally configure two network cards on your network to have the same MAC address, address conflict problems will result and the computers will not be able to communicate on the network. Token Ring cards, however, actually require you to assign a unique MAC address to them before they will work.

The uniqueness of MAC addresses for network devices is ensured by the Institute of Electrical and Electronics Engineers (IEEE), which allocates vendors of networking devices specific blocks of MAC addresses. The first 3 bytes (24 bits) represent the manufacturer of the card, and the last 3 bytes (24 bits) identify the particular card from that manufacturer. Each group of 3 bytes can be represented by 6 hexadecimal digits, forming a 12-digit hexadecimal number representing the entire MAC address. Examples of manufacturer 6-digit numbers include the following:

In order for computers to communicate on a network, frames must be addressed using Layer 2 or MAC addresses. To simplify network communications, however, computers are given Layer 3 addresses instead, which in Transmission Control Protocol/Internet Protocol (TCP/IP) consist of IP addresses. A TCP/IP protocol called Address Resolution Protocol (ARP) is then used to translate Layer 3 addresses into Layer 2 addresses.

Notes

To determine the MAC address of your computer's NIC, use the following commands:

See Also Address Resolution Protocol (ARP) , Ethernet ,frame ,IP address , Open Systems Interconnection (OSI) reference model

Macintosh

A personal computing platform from Apple Computer.

Overview

The Apple Macintosh platform incorporates both a hardware platform and an operating system called MacOS. The original Macintosh hardware platform employed the 680x0 series of microprocessors from Motorola and now uses the Reduced Instruction Set Computing (RISC)-based PowerPC processor developed jointly by Apple, IBM, and Motorola.

The MacOS operating system incorporated a number of innovations into personal computing, including a graphical user interface (GUI) with resizable windows, a mouse, icons for applications, and common activities such as creating Notes and deleting files to the Trashcan. Many of these innovations had been developed earlier by the Stanford Research Institute and the Xerox Palo Alto Research Center (Xerox PARC) and were later licensed by Apple for use by Microsoft Corporation in its Microsoft Windows operating system platform.

History

The Macintosh was first released by Apple in 1984 and quickly became popular with high-end personal computer users. Starting from version 1 in 1984, the MacOS operating system has continued to evolve, as MacOS 8 introduced a new look for the desktop in 1997 and MacOS 9 made significant enhancements in the file system in 1999. The new MacOS X, released in 2001, marked a significant departure from earlier operating system versions by basing the kernel upon a free version of UNIX.

Uses

The Macintosh platform has established a niche for itself in a number of industry sectors, including desktop publishing, graphics, and animation. It is also popular among university professionals. While holding a much smaller share of the market than Microsoft Windows, the Macintosh is a popular alternative desktop platform, but it has made few inroads into the server marketplace. Popular Macintosh platforms include the iMac, iBook, and Power Macintosh versions.

Notes

Using Services for Macintosh, an optional feature of Microsoft Windows 2000, Macintoshes and Intel-based computers can operate together on the same network. Services for Macintosh enables Windows 2000 servers to support Macintosh client machines by emulating an AppleShare server and to provide Macintosh clients with file and print sharing service and AppleTalk routing capability. AppleShare is a file-sharing protocol that is a part of AppleTalk, a suite of networking protocols developed by Apple for its Macintosh computing platform.

For More Information

Visit Apple at www.apple.com

See Also Apple Open Transport ,AppleShare ,AppleShare IP ,AppleTalk ,UNIX ,Windows 2000

MAC layer

Stands for media access control layer, a sublayer of the data-link layer.

See Also media access control (MAC) layer

MADCAP

Stands for Multicast Address Dynamic Client Allocation Protocol, an extension to the Dynamic Host Configuration Protocol (DHCP) for dynamic assignment of multicast addresses.

See Also Multicast Address Dynamic Client Allocation Protocol (MADCAP)

mail client

Software that provides users with an interface for creating, sending, receiving, viewing, and storing e-mail messages.

Overview

Most electronic messaging systems are client/server based, in which users send and receive messages using mail clients and mail servers are used to store and forward mail between users. An example of a client/server messaging system is the combination of Microsoft Exchange Server, a powerful platform for messaging and collaboration, and Microsoft Outlook, the premier messaging client from Microsoft Corporation.

Other popular mail clients include

See Also e-mail , Exchange Server , Simple Mail Transfer Protocol (SMTP)

mailslot

An interprocess communication (IPC) mechanism.

Overview

Mailslots provide connectionless, datagram-based one-to-one or one-to-many communication between processes on different computers running versions of Microsoft Windows. Mailslots are implemented in Microsoft Windows as file system drivers and, therefore,

Mailslots are similar to named pipes, another interprocess communication mechanism. Although named pipes support only point-to-point, bidirectional communications, mailslots support point-to-multipoint, unidirectional communications.

Uses

Mailslots were originally developed for Microsoft OS/2 LAN Manager and have been maintained in later versions of Microsoft Windows for purposes of backward compatibility. They are usually used for broadcast purposes, such as service announcements in the Computer Browser service on Windows 2000 and for the WinPopUp messaging tool in Windows 95 and Windows 98.

See Also interprocess communication (IPC) ,named pipe ,point-to-multipoint ,point-to-point

mail system

Any set of applications used to support e-mail messaging between a group of users within an organization or between users in different organizations.

Overview

E-mail has become as indispensable to modern business as the telephone or fax machine. To implement e-mail within a company, you need to deploy the elements of a mail system. A typical mail system has two basic components:

Architecture

Mail systems generally employ one of two kinds of architecture:

Mail systems can also be distinguished by the address formats they employ. Examples of mail address formats include industry-standard formats such as the Simple Mail Transfer Protocol (SMTP) used on the Internet and the Originator/Recipient (O/R) format developed by the International Telecommunication Union (ITU) for its X.400 messaging standards. Examples of proprietary address formats include the Microsoft Mail version 3.x format and the Lotus cc:Mail format. The following table illustrates examples of these different address formats.

Examples of Address Formats Used in Electronic Messaging

Address Format

Example

SMTP

mitch@northwind.microsoft.com

X.400 O/R

C=US;a=Sprint;p=microsoft; o=northwind;s=Tulloch;g=Mitch;

Microsoft Mail

microsoft/northwind/mitch

cc:Mail

Tulloch, Mitch at northwind

Marketplace

The most popular mail system used in the corporate world today is Microsoft Corporation's Exchange Server platform, which can be deployed with a variety of clients, including

Exchange Server provides a comprehensive messaging solution that supports

Exchange can be deployed as a messaging backbone within a heterogeneous mail system environment and also includes a Migration Wizard that lets you migrate users and mailboxes from legacy and foreign mail systems to Exchange. Consolidation of multiple disparate mail systems into a single messaging system through migration can both simplify administration and reduce costs.

See Also e-mail ,Exchange Server ,Simple Mail Transfer Protocol (SMTP) ,X.400

mainframe

A general term for a high-level, typically large computer that is capable of performing demanding computational tasks.

Overview

Mainframes were the original computing platform developed at the dawn of the computer age in the 1950s and 1960s. Mainframe computers typically operated as centralized computing systems in which dumb terminals were attached through serial connections to a central mainframe computer. Users entered text-based commands into these terminals, and the terminal forwarded the commands to the mainframe, where all the processing was performed. The results of the processing were returned to the terminal and displayed. Modern mainframes typically offer standard 3270 or 5250 terminal communication and also support connections to "smart terminals"-desktop computers that are connected to the mainframe but also have their own computing power.

Mainframe computing established itself in government, industry, and large enterprises as the standard platform for information processing. Purchasing and running mainframes involved considerable expense and required specially trained personnel, so generally only the largest companies and organizations purchased them. Smaller businesses could access mainframe services by submitting batch jobs through remote terminals and paying the company owning the mainframe for processing cycles used.

Vendors of mainframes include IBM, Unisys Corporation, and Hitachi Data Systems, with IBM being the dominant player in today's marketplace. The most widely used mainframe platform is IBM's S/390 system, and a popular midframe platform is IBM's AS/400 system.

Prospects

Mainframe computing went into eclipse in the 1980s and 1990s with the emergence of the PC and client/server computing paradigm, but in recent years, mainframes have been making a comeback as a platform for running Web applications. Many corporations have repurposed older mainframes as Web application platforms, and modern mainframes can run thousands of UNIX or Linux "virtual servers" simultaneously on one physical machine. Once relegated to the role of legacy platform, "big-iron" mainframes are now viewed as fashionable platforms for e-commerce and other Internet-related business paradigms.

Despite the rapid evolution of computing over the last two decades, much of the information processing performed in the government, financial, and industrial sectors still employs traditional mainframes running large COBOL applications to perform functions such as records processing, accounting, and payroll functions.

Notes

Microsoft Host Integration Server can be used as a gateway to provide users on Microsoft Windows networks with access to data stored on IBM S/390 mainframes and AS/400 midframe systems.

For More Information

Learn about the IBM S/390 platform at www.ibm.com/servers/s390/pes.

See Also 3270 ,5250 ,Host Integration Server ,terminal

MAN

Stands for metropolitan area network, which traditionally represents a network spanning a metropolitan area.

See Also metropolitan area network (MAN)

Management Information Base (MIB)

A database of information about a device that is managed using Simple Network Management Protocol (SNMP).

Overview

A Management Information Base (MIB) is a collection of information about managed devices, such as computers, hubs, routers, and switches. MIBs store information about the configuration of these networking components, such as the version of the software running on the component, the Internet Protocol (IP) address assigned to a port or interface, and the amount of available disk space for storage. MIBs thus function as a kind of directory containing information about network settings and resources on managed devices.

Architecture

The SMNP data within an MIB is organized in hierarchical fashion in the form of a tree. The structure of the MIB tree is defined by a number of Internet Engineering Task Force (IETF) standards, including RFCs 1155, 1213, 1514, and 1759.

The generalized MIB tree contains several kinds of branches:

The diagram illustrates the general structure of the MIB tree, which consists of a number of objects represented in ASN.1 notation. Note that a managed device will typically contain only the portion of the entire MIB tree that is relevant to its particular operation. The root of the MIB tree is International Organization for Standardization (iso), followed by Organization (org), Department of Defense (dod), and then Internet (internet). The main public branch is then Management (mgmt), which defines network management parameters common to devices from all vendors. Underneath mgmt is MIB-II (mib-2), and beneath this are branches for common management functions such as system management, host resources, interfaces, and printers. For example, in the diagram we see that the root of the MIB branch that contains objects for SNMP manageable printers is called Printer. In MIB notation, this object is uniquely defined by the text string

.iso.org.dod.internet.mgmt.mib-2.printer

Alternatively, MIB objects are also assigned a numerical label (called an object identifier or OID) that provides a more compact representation of these objects than text strings can provide. Using the numerical values in the diagram, the root Printer object would be uniquely described by

.1.3.6.1.2.1.43

The Private branch of the MIB tree contains branches for large companies and organizations. These are organized under an object called Enterprise, and each vendor has a root branch node under this object. For example, IBM's root branch node is ibm (2), Cisco Systems is cisco (9), Sun Microsystems is sun (42), Microsoft LAN Manager MIB II is lanman (77), and Microsoft Corporation is microsoft (311). Vendors can apply to the Internet Assigned Numbers Authority (IANA) to have specific MIB numbers reserved for their enterprise. Each company or organization has complete authority over what objects will be created within their own branch of the MIB tree and what OIDs will be assigned to each object. All MIB objects must comply with a common definition of SNMP information called Structure of Management Information (SMI), which defines the various data types allowed.

Using Microsoft (.1.3.6.1.4.1.311) as an example of an enterprise, various MIB branches are defined under its root node, as seen here:

.1.3.6.1.4.1.311.1.3 for Dynamic Host Configuration Protocol (DHCP)

.1.3.6.1.4.1.311.1.7.2 for File Transfer Protocol (FTP)

.1.3.5.1.4.1.311.1.7.3 for Hypertext Transfer Protocol (HTTP)

and so on.

Implementation

MIBs are incorporated into special software called SNMP agents that run on SNMP-manageable devices. This is done by using MIB files, which are plain text files constructed using a special format. Once these MIB objects are compiled by the agent software, the device can be managed using a network management system that supports SNMP. You can use SNMP commands to retrieve the value of a MIB object or, in some cases, to change the object's value.

For More Information

Try out the MIB Browser at www.ibr.cs.tu-bs.de/cgi-bin/sbrowser.cgi.

See Also network management ,Simple Network Management Protocol (SNMP)

Management Information Format (MIF)

A standard format for describing hardware and software management information.

Overview

The MIF format and Desktop Management Interface (DMI) specification were developed by the Distributed Management Task Force (DMTF), an organization composed of computer industry leaders whose aim is to lead the development, adoption, and unification of standards for managing systems. MIF can be used in conjunction with DMI to manage networked PCs using systems management software such as Microsoft Systems Management Server (SMS).

Implementation

The MIF database on a computer system is implemented as a collection of plain text files called MIF files. MIF files are text files that are supplied with each manageable hardware or software component installed on the system and that contain information about the attributes of that component. These MIF files provide system management software with configuration information concerning the various hardware and software components on the system. A program called a DMI Service Provider that runs locally and is resident on the system being managed collects and manages the information from MIF files and delivers it to the central management station. You can retrieve and display the information in the MIF database using any standard management interface (MI) utility.

Notes

Microsoft SMS uses six types of MIF files for collecting and storing information in the SMS database:

SMS can also use MIF files to add information to the SMS database about objects such as routers with custom architectures. Note that MIF files differ slightly in syntax between SMS 2 and the earlier SMS 1.2.

See Also Desktop Management Interface (DMI) ,Distributed Management Task Force (DMTF) ,Systems Management Server (SMS)

Management Service Provider (MSP)

A company that manages the IT (information technology) infrastructure for other businesses.

Overview

Management Service Providers (MSPs) are a new breed of service provider that evolved out of the changing application service provider (ASP) marketplace. MSPs offer outsourced IT services that can include

MSPs are particularly appealing to startups, small and mid-sized businesses, and e-commerce companies where rapid time-to-market is essential and where the hiring or training of staff with the required expertise may be too expensive. Instead of purchasing expensive and complex systems management platforms such as HP Openview and IBM Tivoli, outsourcing system and network management tasks to an MSP is a viable option for companies that lack the capital resources or personnel to implement these systems. MSPs can also keep rapidly evolving high-tech companies from being caught in the trap of technological obsolescence by obviating the need to lock in to management systems that may become legacy platforms.

MSPs are part of an evolving breed of xSPs that include ASPs and storage service providers (SSPs). MSPs generally provide their services to client companies on a subscription basis, helping smaller companies avoid the high capital outlay of running their own IT department. MSPs differ from ASPs, whose primary aim is servicing end users on behalf of companies-MSPs instead provide high-level services to businesses and IT departments.

Marketplace

The MSP marketplace is new and evolving rapidly, but some of the players that have made names for themselves include InteQ Corporation, Luminate (now part of EMC Corporation), Manage.Com, Nuclio Corporation, SilverBack Technologies, and Triactive.

For More Information

To learn more about the MSP marketplace and where it is headed, visit the MSP Association online at www.mspassociation.org

See Also application service provider (ASP) ,storage service provider (SSP) ,xSP

Manchester coding

A line coding mechanism used in 10 megabit per second (Mbps) versions of Ethernet.

Overview

Manchester encoding is the technology used since 1979 to convert information into electrical signals for all versions of 10 Mbps Ethernet, including 10Base2, 10Base5, and 10BaseT.

Manchester coding. Example of how Manchester coding works for standard Ethernet.

Manchester coding employs a two-state transition of line voltage to represent one bit of information. In other words, two baud (voltage changes) are used for one bit (piece of information). A binary 0 is represented by a transition from high to low voltage in the time set for transmitting one bit (that is, one "bit time"), while a binary 1 is represented by a transition from low voltage to high. For Ethernet networks, the high voltage is typically +0.85 volts and the low is typically -0.85 volts, making each voltage transition equal to 1.7 volts. This also results in Manchester encoding being balanced for DC operation, which enables signals to pass through devices such as transformers without being corrupted.

Manchester coding enables data to be transmitted between two stations without the need for an extra clocking signal to synchronize communications between the stations. This is possible because voltage transitions take place in the middle of each bit transmission interval, which establishes a timing pattern between stations. The mid-interval voltage changes thus allow the sending and receiving stations to maintain proper synchronization with each other in order to ensure the integrity of the transmission. Because of the extra transition per bit that is used for clocking purposes, Manchester coding is only 50 percent efficient-for example, a 20-megahertz (MHz) bandwidth is required to produce a 10-Mbps data transmission rate.

Advantages and Disadvantages

Manchester encoding has the advantage of being relatively easy to encode (or decode) information. Manchester encoding was selected as the line-coding mechanism for 10 Mbps Ethernet simply because in those days network devices such as hubs, bridges, and network interface cards (NICs) had limited processing capabilities and could easily support Manchester. Another factor in this choice was that coaxial cabling, the original media format used by Ethernet, had a high bandwidth capability and could easily support the 20 MHz signaling required to transmit Ethernet frames at 10 Mbps.

Because Manchester is a relatively wasteful encoding scheme, and with advances in processor power for network devices, newer flavors of Ethernet such as Fast Ethernet and Gigabit Ethernet (GbE) do not employ Manchester.

Notes

Another version of this line coding mechanism called differential Manchester encoding represents binary 0 by a voltage transition at the start of the bit-interval, and binary 1 by no transition at the start of the bit-interval. In both cases, a transition takes place in the middle of the interval for synchronization purposes. Differential Manchester encoding is used for IEEE 802.5 Token Ring networking.

mandatory user profile

A roaming user profile in Microsoft Windows 2000 and Windows .NET Server that the user cannot modify.

Overview

Mandatory user profiles are typically configured by administrators who want to prevent users from modifying their desktops so that administrators can reduce the time they spend troubleshooting modifications or enforcing uniformity because of company policies. Mandatory profiles can also be used for several users who require the same desktop configuration. The administrator can create a mandatory profile and use it for the entire group of users. Changing the mandatory user profile once affects all of the users assigned to it. This simplifies the administrator's job when desktops require upgrading or additional applications.

Because a mandatory user profile is a type of roaming user profile and is therefore stored on a network server, users can access their personal desktop settings from any machine on the network. Mandatory user profiles are configured as read-only, however, so that users cannot permanently change their desktop settings. They can reconfigure their desktop for the duration of the current logon session, but once they log off, their changes are lost.

Implementation

To change a roaming user profile into a mandatory one, simply place the user's roaming profile on the file server where user profiles are stored and rename Ntuser.dat to Ntuser.man. This makes the roaming user profile read-only from the user's perspective.

Notes

Windows 95 and Windows 98 also support mandatory user profiles in conjunction with networks based on Windows NT or Windows 2000.

See Also local user profile ,roaming user profile ,user profile

manual switch

Any manually operated (as opposed to electronically operated) switch used for switching between peripherals or other network devices.

Overview

Manual switches generally cost less than electronic switches that have similar features and capabilities. They are available in a variety of configurations for different kinds of uses and typically have rotary switches on the front and connectors on the back. These connectors could be standard DB9, DB15, RJ-11, RJ-45, RS-232, V.35, BNC, or Centronics connectors.

Manual switch. Two manual switch configurations: a many-to-one switch and an X switch.

Manual switches are typically used in high-security environments in which a user can access a device only by physically switching to it. Some manual switches include key locks that control access.

Types

Examples of different types of manual switches include the following:

See Also switch

MAPI

Stands for Microsoft Application Programming Interface, a Microsoft technology that allows developers to use the Windows messaging subsystem for writing messaging applications.

See Also Messaging Application Programming Interface (MAPI)

mapped network drive

A shared folder on a network that has been associated with an available drive letter on a local PC.

Overview

A mapped network drive enables a shared folder on a remote computer to behave as a logical drive on the local computer. Mapped drives can be created in all versions of Microsoft Windows, typically using procedures such as the following:

Notes

An alternative way of accessing a shared folder on a network server is using the Run command. This method does not require you to use an available drive letter as mapping a drive does. Simply choose Run from the Start menu, and then type the UNC path to the network share you want to access.

See Also Universal Naming Convention (UNC)

markup language

A language used to add information or provide additional meaning to a document.

Overview

Markup languages typically consist of a set of symbols called markup and a set of syntax rules defining how markup should be used. There are generally four types of markup:

History

Markup languages first appeared in the computing world in the 1960s. Soon afterward, attempts were made to unify different types of markup languages, an effort that led in 1986 to the Standardized Generalized Markup Language (SGML) becoming established as an International Organization for Standardization (ISO) standard called ISO 8879. In 1991 SGML was used by Tim Berners-Lee to create a simple markup language called Hypertext Markup Language (HTML) that could be used with the newly developed Hypertext Transfer Protocol (HTTP) to deliver hypertext (linked content) over the Internet. Limitations in HTML led the World Wide Web Consortium (W3C) to propose the new Extensible Markup Language (XML) in the late 1990s. XML is rapidly headed toward becoming the standard language for communicating information over the Internet.

Types

The three kinds of markup language described above have similarities and differences, such as the following:

See Also document type definition (DTD) ,Hypertext Markup Language (HTML) ,Hypertext Transfer Protocol (HTTP) ,International Organization for Standardization (ISO) ,World Wide Web Consortium (W3C) ,XML

master boot record (MBR)

A key data structure on a hard disk.

Overview

The master boot record (MBR) performs an essential role on a hard-disk system of a computer: when a computer boots up, the basic input/output system (BIOS) executes a small portion of code residing in the MBR. This code scans the partition table (found within the MBR) to determine which is the active partition, then scans the active partition to locate the boot sector on this partition, loads the code it found in this boot sector into RAM (random access memory), and then transfers control to this resident code to continue execution or the boot process.

Sometimes the MBR for a disk system can become corrupt-for example, when a system is invaded by an MBR virus. In this case, when you try to boot your system, you may see one of the following messages:

In Microsoft Windows 2000 and Windows .NET Server, you can sometimes repair corruption of the MBR by using the Recovery Console's Fixmbr command.

master browser

A Microsoft Windows 2000 computer that participates in the Computer Browser service.

Overview

The master browser collects and maintains a list of available servers that have shared network resources in its domain. This list, known as the browse list, is created when individual servers announce their presence on the network by sending a server announcement to the master browser.

The master browser also collects and maintains a list of all Windows 2000 domains on the network. If the domain of a master browser spans several subnets, the master browser maintains only the browse list for its own subnet and one master browser will exist for each subnet in the domain.

Backup browsers obtain a copy of the browse list from the master browser. Backup browsers automatically contact the master browser every 15 minutes to request a copy of both the browse list and a list of domains on the network. The backup browser caches this list and distributes it to any client on the network that requests it.

The master browser for a domain is usually the first domain controller installed in that domain.

See Also backup browser ,Computer Browser service ,domain master browser

master domain

In Microsoft Windows NT-based networks, a trusted domain that contains user accounts for the enterprise.

Overview

Master domains are used in enterprise-level implementations of Windows NT to allow user accounts to be centralized and managed in one domain. The master domain is typically found at the company headquarters, while resource domains are implemented at branch offices. A trust relationship is established between the resource domains and the master domains, in which each resource domain trusts the master domain. Users who log on to their computers at headquarters automatically log on to the master domain to validate their credentials. Users at branch offices can log on to either their local resource domain or the trusted master domain, but they must choose the master domain because all user account information resides there. In a wide area network (WAN), one or more backup domain controllers (BDCs) belonging to the master domain are installed at each branch office to facilitate local logons and to prevent logon traffic from being routed over slow WAN links to headquarters.

Notes

Many companies use the Windows NT master domain model to administer users and groups in master domains and resources in second-tier (or resource) domains. Because the majority of reasons to use the Windows NT master domain model disappear when domains are migrated to Active Directory directory service in Windows 2000 and a domain tree is established, these companies might choose to dissolve existing second-tier domains into organizational units (OUs) in the master domains.

See Also Active Directory ,domain (Microsoft Windows) ,organizational unit (OU) ,resource domain ,trust

master index

The final version of an index generated by Microsoft Indexing Service.

Overview

The indexing process first creates word lists, which are merged into shadow indexes and then become one master index. Unlike word lists, which reside in volatile memory, the shadow indexes and the master index are persistent indexes that are stored on disk. The process by which shadow indexes are combined into a master index is called a master merge. A master merge combines multiple shadow indexes with any old master index to create a new master index. The new master index is the most efficient form for the index of a corpus of documents being indexed by the Indexing Service.

The master index stores the indexing data in a highly compressed format in a structure called the catalog. The compressed nature of the master index provides the most efficient data structure for storing information and for issuing queries against the corpus, but the process of generating a master index is resource intensive and is generally performed by the Indexing Service when load on the server is low. However, the administrator can also force a merge to update the master index after new documents are added to the corpus being indexed.

See Also catalog

master name server

A name server in the Domain Name System (DNS) that has authority over a given zone.

Overview

A master name server can be either a primary name server or a secondary name server. If it is a secondary name server, the master name server must obtain its own resource records from another master name server. If it is a primary name server, it contains the master copy of the resource records for its zone of authority.

Master name servers provide read-only copies of their DNS database to secondary name servers. A secondary name server must obtain and update its resource records from a master name server. The process by which these records are obtained and updated between name servers is called a zone transfer.

Notes

When you configure zone transfers between name servers, be sure that the start of authority (SOA) record for the secondary name server is correctly configured with the name of the master name server from which it will obtain its resource records. No configuration is required with the master name server itself.

See Also caching-only name server ,Domain Name System (DNS) ,name server ,primary name server ,secondary name server

master station

Any network device that controls the operation of other network devices.

Overview

The term master station is typically used in mainframe/terminal networking. For example, in a point-to-point connection between an IBM controller and a 3270 terminal, the controller is the master station while the terminal is a slave station. Or, in a multipoint circuit, one of the devices might function as a master station while all the rest are slaves.

Typically, the master station polls all the slave stations regularly in order to determine whether a slave wants to initiate communication with the master. This is different from a peer multipoint circuit such as an Ethernet network, where each station contends for control of the media.

Notes

Another scenario where the term master station sometimes is used is in Token Ring networks, where one of the devices can function as a master station and be responsible for detecting and replacing tokens lost by other stations.

See Also 3270 , Ethernet , terminal, Token Ring

matrix switch

A multiport switch that supports any-to-any connectivity.

Overview

The term matrix switch originates from a mathematical structure called a matrix, which is a two-dimensional structure with N rows and M columns representing N times M values. Likewise, a matrix switch with N input ports and M output ports has N x M switching possibilities. Typical configurations for these switches include 4 x 4 and 4 x 8 matrix switches, in which any input device can be connected to any output device by operating the switch.

Matrix switch. Using a matrix switch to connect four printers to four computers.

Types

A typical matrix switch would be an electronic switch that has multiple ports and that can be controlled using a keypad or some other front-panel interface on the switch. Another type of switch called a code-operated switch is basically a switch that can be operated using character codes embedded in the data stream sent from user workstations. Code-operated matrix switches can thus be operated remotely from computer terminals by entering ASCII commands into a text window.

Uses

Matrix switches are typically used in serial connections such as RS-232 to allow multiple computers to share a group of peripherals such as printers, modems, and other devices. A high-end matrix switch may consist of a chassis into which ports can be added as required to support different interfaces and may be packaged either as a desktop unit or a standard 19-inch (48-centimeter) rackmount enclosure.

A popular use of matrix switches is for Keyboard Video Mouse (KVM) switches, which incorporate matrix switching technology to allow multiple computers to access multiple monitors, keyboards, and mice simultaneously. Although low-end KVM switches may allow a single keyboard, monitor, and mouse to be switched between two, four, or eight different computers, higher-end KVM switches are essentially matrix switches that allow X keyboard/ monitor/mouse combinations to actively control Y different computers, with high-end configurations supporting up to 128 computers and 128 different users. Examples of vendors providing KVM matrix switches include ITM Components, Network Technologies, and Lightwave Communications.

Also available are matrix switches that support Small Computer System Interface (SCSI) and parallel communications interfaces. For example, a 4 x 2 SCSI matrix switch could allow two hosts (computers) to switch between four different SCSI drive systems (such as redundant array of independent disks [RAID] arrays). SCSI matrix switches generally allow multiple hosts to access multiple drive systems concurrently, subject to an overall maximum throughput and to the 15 device SCSI limit. An example of a vendor of SCSI matrix switches is Avax International, which provides a modular SCSI matrix switch that can be customized for different host/device combinations.

Notes

The term matrix switching also describes the switching technology at the center of an Asynchronous Transfer Mode (ATM) network that functions by supplying needed bandwidth for end-to-end sessions. ATM matrix switching avoids contention by end stations on the network.

See Also code-operated switch ,Keyboard Video Mouse (KVM) switch ,parallel transmission ,RS-232 ,serial transmission ,Small Computer System Interface (SCSI) ,switch

MAU

Stands for Multistation Access Unit, a wiring concentrator (passive hub) used in Token Ring networks.

See Also Multistation Access Unit (MAU or MSAU)

MBone

An experimental portion of the Internet that is used to test multicasting technology.

Overview

The MBone platform was established to test and develop technology and software for multicasting audio and video information over the Internet. Although the MBone came into existence in 1992, Internet Protocol (IP) multicasting actually dates back to 1988, when the first multicast tunnel was established between Stanford University and BBN Technologies (now part of Verizon Communications). The MBone initially joined together 40 academic and research networks in four different countries but expanded rapidly to include thousands of networks in dozens of countries and regions.

The MBone is essentially a virtual multicasting network lying on top of the Internet and implemented using routers configured to support IP multicast transmission. These multicast-enabled routers can forward IP packets through nonmulticast routers using tunneling technologies that encapsulate multicast IP packets into unicast packets that the routers can understand. MBone IP addresses used for multicasting of audio and video information have been allocated by the Internet Assigned Numbers Authority (IANA) as those in the range 224.2.y.z, which is a subset of the class D group of IP addresses ranging from 224.0.0.0 to 239.255.255.255.

Notes

The name MBone was selected because the network represents a multicast backbone; that is, a backbone network for multicasting.

See Also Internet ,

MBR

Stands for master boot record, a key data structure on a hard disk.

See Also master boot record (MBR)

MBS

Stands for Mobile Broadband System, a proposed fourth- generation (4G) mobile cellular communication system.

See Also Mobile Broadband System (MBS)

MBSP

Stands for Multitenant Broadband Service Provider, another name for a building-centric local exchange carrier, a telecommunications carrier focused on the Multitenant Unit (MTU) market.

See Also Multitenant Broadband Service Provider (MBSP)

MCP

Stands for Microsoft Certified Professional, a family of technical certifications relating to proficiency in working with Microsoft products.

See Also Microsoft Certified Professional (MCP)

MCS

Stands for Microsoft Consulting Services, a service from Microsoft Corporation that provides direct assistance to large enterprises for the planning, deployment, and support of solutions based on Microsoft technologies.

See Also Microsoft Consulting Services (MCS)

MDA

Stands for media-dependent adapter, a standard for modular Ethernet switches.

See Also media-dependent adapter (MDA)

MDAC

Stands for Microsoft Data Access Components, a set of Microsoft technologies that provide access to information stored in a broad range of data sources.

See Also Microsoft Data Access Components (MDAC)

MDHCP

Stands for Multicast Dynamic Host Configuration Protocol, the former name for what is now called Multicast Address Dynamic Client Allocation Protocol (MADCAP).

See Also Multicast Address Dynamic Client Allocation Protocol (MADCAP)

media

Anything over which signals can be transported.

Overview

In computer networking, the term media refers generally to the types of cabling over which data is transmitted. Types of cabling for networking and telecommunications include copper cabling and fiber-optic cabling. Copper cabling generally comes in two main forms:

Fiber-optic cabling is used mainly in campus backbones and for long-haul telecommunications. At one time, it was expected that fiber would supplant copper, but improvements in switching technologies has enabled copper-based networks to support gigabit speeds and higher.

Another type of media is not really media at all: wireless networking. This type of networking transports signals through empty space using radio and microwave frequency electromagnetic radiation traveling at the speed of light. Although most wireless networking signals travel near the Earth's surface and hence go through air, this air is not considered a medium-electromagnetic waves do not require a medium in which to travel. Wireless networking is sometimes said to employ "unguided media" in contrast to the "guided media" provided by wires and cables. Another way of saying this is to contrast "wireline" (guided) networks with wireless (unguided) ones.

Notes

Another common use of the term media is to refer to a combination of cabling and topology used for different types of Ethernet networks. Thus, for example, 10BaseT refers to a star topology using unshielded twisted-pair (UTP) cabling, and 10Base2 refers to a bus topology using thinnet cabling. You can loosely refer to these examples as 10BaseT and 10Base2 Ethernet media.

See Also cabling ,coaxial cabling ,fiber-optic cabling ,twisted-pair cabling ,wireless networking

media access control (MAC) layer

A sublayer of the data-link layer.

Overview

The Project 802 specifications of the Institute of Electrical and Electronics Engineers (IEEE) subdivide the data-link layer (Layer 2) of the Open Systems Interconnection (OSI) reference model into two sublayers:

The function of the MAC layer is to decide which station on the network is allowed to use the media at any given moment. The MAC layer is therefore responsible for implementing the particular media access control method on a network and thus distinguishes different networking architectures such as Ethernet and Token Ring. The table illustrates the different types of MAC layers defined by Project 802.

MAC Layers Defined by Project 802

Project 802 Committee

Type of LAN

Channel Management Algorithms Defined

802.3

Carrier Sense Multiple Access with Collision Detection (CSMA/CD)

Binary backoff after collision detection, framing, error detection

802.4

Token bus LAN

Token passing, priority scheme, framing, error detection

802.5

Token ring LAN

Token passing, priority scheme, framing, error detection

As part of the OSI protocol stack, the MAC layer receives framed data from the LLC layer immediately above it, which is media independent, and reframes the data, adding a source and destination physical address or MAC address to the frame for transmission on the medium. The MAC layer is also responsible for making sure that data is delivered without errors to layers above it.

See Also 802.3 , 802.4 ,802.5 ,Carrier Sense Multiple Access with Collision Detection (CSMA/CD) ,data-link layer ,Ethernet ,Institute of Electrical and Electronics Engineers (IEEE) ,logical link control (LLC) layer , Open Systems Interconnection (OSI) reference model, Project 802, Token Ring

media access control method

Any method for allowing multiple stations to transmit signals over a network without conflict.

Overview

If two computers simultaneously place signals on a network cable, a collision can occur. Collisions can result in data being lost or corrupted. The solution is to provide a media access control method for the network. Media access control methods act as traffic lights by permitting the smooth flow of traffic on a network and either prevent collisions entirely or provide a graceful way of dealing with them when they occur.

Media access control methods are implemented at Layer 2, the data-link layer, of the Open Systems Interconnection (OSI) reference model. Specifically, media access control methods are the responsibility of the media access control (MAC) layer, one of two sublayers of the data-link layer. Four main media access control methods are used in local area networks (LANs):

Implementation

In real-life networking devices such as network interface cards, switches, and routers, media access control methods are implemented using a MAC algorithm. Although the algorithms for Ethernet and Token Ring networks are publicly defined by the IEEE standards described above, those used in full-duplex Ethernet are usually patented by the companies that developed them and are hard-coded into application-specific integrated circuits (ASICs) to achieve the best possible performance. For example, in full-duplex Fast Ethernet and Gigabit Ethernet (GbE) networking, the most popular MAC layer is that designed by Alcatel, from which most switch and router vendors lease their MAC technology. The Alcatel MAC is used in more than half of all GbE ports on the market today and in more than 80 percent of all Fast Ethernet ports. Network engineers should note that mixing switching equipment from vendors that use different MACs can sometimes lead to interoperability problems.

See Also 802.3 , 802.4 ,802.5 ,802.11a ,802.11b ,802.12 ,AppleTalk ,Carrier Sense Multiple Access with Collision Avoidance (CSMA/CA) ,Carrier Sense Multiple Access with Collision Detection (CSMA/CD) ,demand priority ,Fiber Distributed Data Interface (FDDI) ,Institute of Electrical and Electronics Engineers (IEEE) , Project 802, Token Ring

media converter

A device that connects two different networking media.

Overview

Media converters are commonly used for Ethernet networks where each type of Ethernet has different specifications for the media that can be used. For example, 10Base2 and 10Base5 use different media types (thinnet and thicknet cabling, respectively) and 10BaseT employs unshielded twisted-pair (UTP) cabling. Media converters allow these different types of cabling to be joined together transparently.

Implementation

Media converters come in many types, with many different kinds of connectors. They range from simple two-port small boxes for connecting two different cables to expensive rack-mounted chassis units with 8 or 16 modular ports that support a wide range of configuration options. Some media converters also include bridging functionality and can be deployed wherever bridges or switches are used.

Media converter. Using a media converter to connect unshielded twisted pair (UTP) and fiber optic cabling.

Uses

Common uses of media converters in networks include

Media converters can often be valuable for companies that have a legacy thicknet or thinnet Ethernet local area network (LAN) but want to take advantage of Fast Ethernet switching technology. Instead of rewiring your entire LAN to unshielded twisted-pair (UTP) cabling, which is the standard cabling type for 100BaseT Ethernet, you can utilize some of your existing cabling by connecting your coaxial cabling to a media converter and then connecting the media converter to the Ethernet switch using UTP patch cables.

Media converters are also commonly used to join copper LANs together using long runs of fiber. Canoga Perkins Corporation offers its EdgeAccess Gigabit Ethernet Media Converter for connecting Gigabit Ethernet (GbE) backbone switches using fiber-optic cabling. Products such as this can be used to build custom metropolitan area networks (MANs) using dark fiber leased from telcos and are often used by service providers that are building metropolitan Ethernet backbone networks.

See Also 10Base2 , 10Base5 ,10BaseT ,100BaseT ,coaxial cabling ,Ethernet ,Fast Ethernet ,fiber-optic cabling ,Gigabit Ethernet (GbE) , thicknet, thinnet, unshielded twisted-pair (UTP) cabling

media-dependent adapter (MDA)

A standard for modular Ethernet switches.

Overview

Ethernet switches come in two types:

Modular Ethernet switches generally consist of a chassis with two to four media-dependent adapter (MDA) slots on the front panel. To customize these switches for your network, you simply insert MDA modules into the slots to provide the media configuration you require. A single MDA module might provide one of the following typical port configurations:

By selecting and inserting different MDAs into a modular Ethernet switch, you can quickly reconfigure complex networks. Bridging between the different types of media is performed internally by the chassis circuitry inside the switch.

See Also 10BaseT ,100BaseFX ,100BaseT ,100BaseTX ,Ethernet ,Fiber Distributed Data Interface (FDDI) ,fiber-optic cabling ,unshielded twisted-pair (UTP) cabling

Megaco

An emerging framework for connecting voice, fax, and videoconferencing systems to packet-switching networks.

Overview

Megaco is a standard developed jointly by the International Telecommunication Union (ITU) and the Internet Engineering Task Force (IETF). Megaco is designed to support convergence of telephone and data services and is especially targeted toward the Internet Protocol (IP) telephony sector to ensure interoperability between gateways and controllers supplied by different vendors. The name Megaco is short for Media Gateway Control, and the ITU standard for this framework is H.248.

Architecture

Integrating circuit-switched systems such as the Public Switched Telephone Network (PSTN) and packet- switched systems such as the Internet are complicated by the fact that the PSTN actually consists of two separate networks:

In a typical IP telephony system, the IP network interfaces with the two PSTN networks using two different kinds of devices:

Megaco's role in the IP telephony framework is to provide a standard protocol for communication between MGs and MGCs. Megaco allows MGCs to control MGs and enables the remote control of connection-aware devices across signaling domains. Megaco can be implemented on a wide range of devices, including

Prospects

Megaco is intended as an industry standard, vendor- independent protocol for MG/MGC communication and is expected to displace the proprietary Media Gateway Control Protocol (MGCP) defined in RFC 2705, which has established itself as a de facto standard but which lacks some of Megaco's advanced capabilities, such as peer-to-peer device control (MGCP supports only master/slave control). Megaco also replaces earlier protocols, including Internet Protocol Device Control (IPDC) and Simple Gateway Control Protocol (SGCP).

See Also Digital Subscriber Line Access Multiplexer (DSLAM) , H.323 ,IP telephony , Public Switched Telephone Network (PSTN), Voice over IP (VoIP)

member server

A server that does not perform logon authentication.

Overview

In Microsoft Windows 2000-based networks, a member server is any server that is not a domain controller. A member server, therefore, does not contain a copy of Active Directory directory service. Member servers do participate in domain security, however, by allowing folders and other resources on them to be shared with permissions assigned to different users and groups. However, member servers cannot be used to authenticate the credentials of users so they can log on to the network-domain controllers are required to perform this action.

Member servers are usually used for dedicated network purposes such as the following:

Notes

In Windows 2000, you can change a member server into a domain controller by running the Active Directory Installation Wizard.

See Also Active Directory ,domain controller

mesh topology

A network topology that has redundant data paths between nodes.

Overview

A mesh network (a network that has a mesh topology) is any network in which information typically has more than one route it can take between any two end stations. Meshed topologies are commonly used in backbone networks to provide fault tolerance-if a wire, hub, switch, or other component fails, data can always reach its destination by traveling along an alternate path. Meshed networks are common in older, routed Internet Protocol (IP) internetworks and prevent the failure of a single router from bringing down a portion of the network.

Mesh topology. Example of mesh topology for a routed internetwork.

There are two basic types of mesh networks:

See Also bus topology ,fault tolerance ,internetwork ,ring topology ,router ,star topology

message

Information sent between two entities.

Overview

The common definition of a message in computer networking is a communication sent by one user to another user by means of an e-mail system such as the Simple Mail Transfer Protocol (SMTP) mail system that is used on the Internet. More generally in computer networking, a message is any grouping of information at the application layer (Layer 7) of the Open Systems Interconnection (OSI) reference model that is exchanged between applications for various purposes. These Layer 7 messages can contain control information or report errors, or they can support the running of a network-aware application in some other fashion.

In Microsoft Message Queuing (MSMQ) terminology, a message is a unit of information or data that is sent between two applications, usually hosted on different computers on a network.

See Also application layer ,e-mail ,Internet ,Open Systems Interconnection (OSI) reference model ,Simple Mail Transfer Protocol (SMTP)

message digest (MD) algorithms

A series of hashing algorithms used in cryptography.

Overview

Message digest (MD) algorithms are commonly used in cryptography to generate digital signatures for ensuring the integrity and authenticity of encrypted messages. The term message digest refers to a short string or hash value of fixed length that is computed from the longer variable-length message being hashed by the algorithm.

Types

The important message digest algorithms include MD2, MD4, and MD5, all of which produce a 128-bit hash value. The details of these different algorithms are as follows:

See Also cryptography , digital signature ,encryption ,hashing algorithm , Point-to-Point Protocol (PPP)

Message Tracking Query Protocol (MTQP)

An emerging protocol for tracking e-mail messages over the Internet.

Overview

Message Tracking Query Protocol (MTQP) is a proposed standard from the Internet Engineering Task Force (IETF) to enable administrators to track the route that Simple Mail Transport Protocol (SMTP) messages take as they travel across the Internet. MTQP enables a message's sender to determine the specific host and router hops taken by SMTP messages, regardless of whether they reach their destination or not. MTQP thus provides the capability for determining whether messages have actually reached their intended recipients or where along the way they may have been prevented from reaching their destination.

MTQP solves a problem inherent with SMTP: the current best-effort delivery model of this protocol means that there is no guarantee that messages will reach their recipients and no way of knowing if they are delayed or lost. In this respect, X.400 messaging systems based on the X.400 recommendations developed in the 1980s by the International Telecommunication Union (ITU) have had a distinct advantage over SMTP, namely, messaging accountability. This accountability enables X.400 systems to be used for important purposes such as sending contracts and other legal documents-if the message is not received, the sender knows it and can act accordingly. As a result, X.400 continues to be used in situations where message integrity is vital, as in certain international and government organizations such as the North Atlantic Treaty Organization (NATO). Until now, however, SMTP lacked the essential accountability to make it a suitable medium for transmission of this kind of information. With a combination of MTQP, encryption, and digital signatures, the Internet may soon become the de facto medium for secure electronic transmission of legal information such as contracts, agreements, invoices, and bank statements.

MTQP was designed with input from AT&T Labs, MessagingDirect (now part of ACI Worldwide), and the Sendmail Consortium. The protocol operates in conjunction with SMTP Service Extensions (ESMTP), enabling the exact path that a message takes as it travels between hosts and across routers over the Internet.

Prospects

MTQP will be supported by upcoming versions of Sendmail and will likely be supported by other Internet mail platforms as well. MTQP's full value will only be realized, however, if it becomes widely implemented across the Internet and by major Internet service providers (ISPs). However, MTQP can be implemented on a limited basis between companies that frequently exchange e-mail and can alert them to problems with their messaging communications.

See Also ESMTP ,Internet Engineering Task Force (IETF) ,Simple Mail Transfer Protocol (SMTP) ,X.400

Messaging Application Programming Interface (MAPI)

A Microsoft technology that allows developers to use the Windows messaging subsystem for writing messaging applications.

Overview

The Messaging Application Programming Interface (MAPI) provides a generic programming interface for making Windows applications mail-enabled and standardizes how messages are handled by messaging applications. MAPI also provides a general messaging subsystem built into the Windows operating system that can function with any message transport mechanism. The messaging subsystem is thus similar to the printing subsystem that allows Windows to interface with any kind of print device and perform standard printing functions.

Implementation

MAPI is a kind of middleman between the messaging application running on the computer and the underlying messaging services. The client interface for accessing these services through MAPI is the same whether the services are local area network (LAN)-based messaging services, e-mail services, fax services, or any other messaging service. MAPI allows standard function calls to access general messaging functions such as sending, receiving, and reading messages in a uniform fashion regardless of the messaging subsystem used. MAPI provides two sets of interfaces: the client-side application programming interfaces (APIs) called by the messaging application, and the service-provider interfaces that the Windows messaging subsystem uses to connect with different messaging-handling systems.

Microsoft Outlook is an example of a MAPI-enabled client, and Microsoft Exchange Server supports MAPI messaging.

See Also application programming interface (API) ,Exchange Server ,Outlook

metabase

A database for storing configuration settings for Microsoft Internet Information Services (IIS).

Overview

The metabase is a hierarchical, memory-resident database that is used to store IIS configuration information. Prior to version 4 of IIS, such information was stored mainly in the Microsoft Windows NT registry. The metabase was introduced in IIS 4 as a more flexible and expandable repository than the registry, and being memory-resident, it also provides better performance than the disk-based registry. Some settings for IIS are still stored in the registry, however, for backward compatibility with older versions of IIS.

The metabase is stored as the file Metabase.bin in the \WINNT\system32\inetsrv directory on a Windows 2000 machine running IIS.

Notes

The Microsoft Internet Information Server Resource Kit includes a number of administrative scripts that you can use to back up and restore the metabase.

See Also Internet Information Services (IIS) ,registry

metadirectory

A tool for combining diverse data sources into a single directory.

Overview

One of the realities of enterprise networks is heterogeneity, and this is particularly apparent in where and how an enterprise stores its data. A large enterprise might have several directory services such as Microsoft Active Directory directory service, Novell Directory Services (NDS) eDirectory, a homegrown Lightweight Directory Access Protocol (LDAP) directory, user account directories such as those found in Microsoft Windows NT and Novell NetWare, database platforms such as Oracle and Microsoft SQL Server, and messaging applications such as Lotus cc:Notes and Microsoft Exchange Server 5.5 that use their own proprietary directories. The task of a metadirectory is to integrate these various data sources into a single hierarchical directory that can be managed and queried as easily as if a single directory service were used across the enterprise.

Implementation

A metadirectory is really an integration strategy rather than a specific set of tools. Important decisions have to be made when thinking about integrating disparate directories together in the enterprise. For example, will one directory be chosen as the master and all others be slaves with respect to it? If this is the case, the directory chosen must be sufficiently scalable, robust, and extensible to enable it to integrate with all the other kinds of directories in the enterprise. In some situations, it may make sense for several directories to operate as masters with respect to different kinds of information. At other times, tools may be selected to enable several directories to operate as peers with each other, but this is rarer.

Another important question to decide is the level of integration to be achieved. For example, will updates be propagated across different directories or will queries simply be distributed across them instead? This question boils down to two basic ways of implementing integration of different directories: synchronization and brokering.

Marketplace

Microsoft MetaDirectory Services (MMS) from Microsoft Corporation is a metadirectory product originally developed by Zoomit Corporation that can be used for integrating different directories together. MMS uses a core metadirectory service called Metaverse that communicates with other directories such as Microsoft Active Directory using management agents. MMS agents are available for a wide range of enterprise applications, including enterprise resource planning (ERP) and customer relationship management (CRM) applications from different vendors.

Another metadirectory product is Novell's new DirXML platform, which uses XML-based connectors to integrate Novell eDirectory to directory services from other vendors and to various messaging platforms. Connectors will be available for a wide range of enterprise applications, including those from SAP and PeopleSoft, X.500 directory services, and databases that support the Open Database Connectivity (ODBC) and Java Database Connectivity (JDBC) platforms.

Prospects

Although LDAP has securely replaced the complex and unwieldy X.500 directory recommendations from the International Telecommunication Union (ITU), the emergence of Extensible Markup Language (XML) has complicated the directory landscape somewhat. Most commercial directories, such as Microsoft Active Directory and Novell eDirectory, are LDAP-compliant and support LDAP application programming interfaces (APIs) that make it fairly simple to enable these directories to communicate with each other. Other applications that have non-LDAP data stores can often be integrated with LDAP directories by writing custom middleware that translates application-specific APIs into LDAP APIs. With the emergence of XML, however, new strategies are being tried for storing enterprise information, and an initiative called Directory Services Markup Language (DSML) is being developed to further support the integration of vendor-specific directories. Enterprises should not view metadirectories as a panacea for directory interoperability problems but as a workaround for lack of planning and unstructured growth of enterprise information sources, and these products are sometimes little more than veiled attempts by vendors to persuade customers to migrate to the vendor's own directory platform.

See Also Active Directory ,directory ,Directory Service Markup Language (DSML) ,Java Database Connectivity (JDBC) ,Lightweight Directory Access Protocol (LDAP) ,Novell Directory Services (NDS) ,open database connectivity (ODBC) ,X.500 ,XML

metasearching

Integrating multiple search engines into a single search entity.

Overview

Search engines such as Google, AltaVista, and Yahoo! are popular tools for users to find helpful information on the Internet. Metasearching simply means integrating several of these tools using a single search interface. Metasearching can employ several architectures:

See Also Internet

metropolitan area network (MAN)

Traditionally, a network spanning a metropolitan area.

Overview

Metropolitan area network (MAN) is a concept that can mean several things depending on the context. For example, a MAN could be

The first definition on this list is the one most familiar to enterprise network architects. To build a WAN consisting of LANs in different geographic locations, enterprises lease MAN services from their local telco. These MAN services thus enable LANs to be joined together into a WAN, and the MAN thus functions as the demarcation point between LAN and WAN. Common MAN services offered by telcos include Integrated Services Digital Network (ISDN), T1 and T3 lines, and Digital Subscriber Line (DSL) services. Enterprises typically interface their LANs with the telco MAN by deploying an access device (also called a WAN switch or access multiplexer) in the basement of their equipment room. The access device then aggregates company LAN traffic and backhauls it from the customer premises to the Point of Presence (POP) of the telco, where it typically traverses the metropolitan area over a dual SONET ring.

Prospects

A new development is the emergence of metropolitan Ethernet service providers, which compete with traditional telcos for hauling data traffic by building their own fiber MANs and provisioning fiber to businesses in downtown areas. Metropolitan Ethernet is cheaper than traditional telco WAN services that rely on ATM over SONET, is easier to deploy and manage, and is fast becoming a viable alternative for the MAN demarcation point between LAN and WAN.

See Also ATM over SONET , local area network (LAN) , wide area network (WAN)

metropolitan Ethernet

A service that uses Gigabit Ethernet (GbE) to provision metropolitan areas with high-speed data services.

Overview

Metropolitan Ethernet represents a new kind of data service now offered in many dense urban areas. Traditionally wide area network (WAN) and metropolitan area network (MAN) services have been provisioned by telcos, particularly by Incumbent Local Exchange Carriers (ILECs) that own the local loop telephone wiring connecting customer premises to their services. Telecommunication services offered by telcos have included dial-up Internet access, Integrated Services Digital Network (ISDN), T1 and fractional T1 lines, and recently Digital Subscriber Line (DSL). The monopoly enjoyed by telcos as a result of local loop ownership has enabled them to maintain high prices for many of these services. For example, a T1 line typically costs around $1,000 a month and provides only 1.5 megabits per second (Mbps) bandwidth, but a T3 line offers a much higher bandwidth of 45 Mbps but at a correspondingly higher price of tens of thousands of dollars per month, which is beyond the budget of most businesses except the largest enterprises and large service providers. DSL is a relatively new technology that offers multimegabit data speeds, and competition with cable modem technologies has kept DSL prices low for residential customers, but the absence of cable television infrastructure buildout in business parks and downtown areas has meant that DSL prices for business customers can be kept artificially high.

The result of all this is that the MAN has become the bottleneck between the corporate local area network (LAN) and the WAN. Although LAN speeds have risen to 100 Mbps Fast Ethernet and higher and WAN speeds within the Synchronous Optical Network (SONET) backbones of carriers is also 100 Mbps or higher, companies trying to join their LANs into a WAN using traditional T1 lines find that these lines have become the bottleneck in the equation and that upgrading to T3 is too costly a solution.

Enter the metropolitan (or metro) Ethernet service providers, companies who are building their own separate MANs to compete with the existing SONET rings owned by telcos. These new service providers are building their MANs using GbE switching gear, laying their own fiber backbones in dense urban areas to connect their points of presence (POPs) with industrial parks, campuses, and downtown multitenant units (MTUs). The demarcation point at the customer premises now becomes as simple as an RJ-45 connection, and customers can simply use a Category 5 (Cat5) patch cord to connect their Ethernet LAN to their carrier's GbE backbone to build a WAN or receive high-speed Internet access. Other services supported by some metro Ethernet providers include high-speed connectivity to storage area networks (SANs), Web and application server hosting, and multicast Internet Protocol (IP) video services.

Implementation

Metro Ethernet is made possible by the use of new carrier-class GbE switches that can be deployed at either the service provider's POP or the basement of the customer's MTU, depending on the scenario. At the provider's end, these switches interface with telco SONET rings to provide connectivity for long-distance WAN links; within a downtown area of a city, however, the metro Ethernet provider maintains end-to-end Ethernet connectivity for customers without the need of telco SONET services.

Metro Ethernet service providers may either lay their own fiber downtown or lease dark fiber from other carriers. To get the most out of this fiber, providers are starting to employ Wavelength Division Multiplexing (WDM), which enables as many as eight full-duplex GbE channels to run over a single fiber pair. Dense Wavelength Division Multiplexing (DWDM) offers the potential of squeezing even greater bandwidth out of individual fiber strands.

With GbE now supporting distances up to 150 miles (240 kilometers) over single-mode fiber, metro Ethernet can be deployed across even the largest cities to provide a high-speed alternative to SONET for data transport. The emergence of 10 Gigabit Ethernet (10 GbE) promises to provide even greater capacity for this service.

Advantages and Disadvantages

Metro Ethernet offers a number of advantages over traditional SONET rings for building WANs:

The main disadvantage of metro Ethernet is availability: for the foreseeable future, this service will probably be available only in the downtown sections of large urban areas. Another disadvantage is that although dual SONET rings deployed by telcos are self-healing (typically in less than 50 milliseconds), GbE networks do not have such built-in recoverability features. Furthermore, while ATM over SONET is a cell-based connection-oriented technology that is optimized for voice but can also carry data easily, Ethernet is a connectionless packet-switched technology that suffers unpredictable delay and jitter, factors that have little effect on data traffic but which make it difficult to transport voice traffic (unless bandwidth is far below saturation point). Finally, SONET networks traditionally offer five-nines (99.999 percent) uptime, which is typically an order of magnitude better than the best metro Ethernet networks available today.

Marketplace

Metro Ethernet service providers offer a wide range of services from LAN-LAN interconnection within urban areas to high-speed Internet access and even Voice over IP (VoIP) services. They can provision a single fiber link to an MTU in such a way that each client within the building has its own secure connection and is billed separately according to usage. Popular service providers in the metro Ethernet marketplace include Yipes Communications, Telseon, FiberCity Networks, Cogent Communications, Everest Broadband Networks, Intelli- Space, XO Communications, and many others. Although some of these providers offer services only in a few urban centers, many of them are scaling out to offer services nationally in large cities across the United States. Some metro Ethernet service providers specialize only in high-speed Internet access or WAN connectivity, but others offer a full range of services comparable to telcos.

Carrier-class GbE switches for metro Ethernet rollouts are now available from a number of vendors, including Extreme Networks, Foundry Networks, Cisco Systems, Nortel Networks, and Riverstone Networks.

Prospects

Just as GbE has won out over competing LAN backbone technologies such as Fiber Distributed Data Interface (FDDI) and Asynchronous Transfer Mode (ATM), the same thing may happen in the WAN, at least for data services. ATM over SONET still offers a much higher Quality of Service (QoS) for voice communications, but the simplicity and price advantages of end-to-end Ethernet in the WAN make it the leading technology as far as data services are concerned. Unfortunately, metro Ethernet is likely to remain exactly that-Ethernet restricted to metropolitan areas where dense clusters of MTUs make laying fiber a cost-effective decision for service providers. Metro Ethernet is unlikely to be available in the foreseeable future in smaller urban or rural areas, which must continue to rely on telco services as their only option. Nevertheless, metro Ethernet is having a profound impact on the enterprise WAN cost model and may even eat into the traditional voice market of telcos as VoIP technologies become more standardized and widely deployed.

See Also Asynchronous Transfer Mode (ATM) , ATM over SONET ,dense wavelength division multiplexing (DWDM) ,Digital Subscriber Line (DSL) ,Fiber Distributed Data Interface (FDDI) ,Gigabit Ethernet (GbE) ,Incumbent Local Exchange Carrier (ILEC) ,Integrated Services Digital Network (ISDN) ,local area network (LAN) , storage area network (SAN), Synchronous Optical Network (SONET), T-carrier, Voice over IP (VoIP), wide area network (WAN)

MExE

Stands for Mobile Execution Environment, an emerging standard for running Java applications on mobile phones.

See Also Mobile Execution Environment (MExE)

MFC

Stands for Microsoft Foundation Classes, a set of object-oriented interface to the Microsoft Windows application programming interface (API).

See Also Microsoft Foundation Classes (MFC)

MFR

Stands for Multilink Frame Relay, a new frame relay aggregation technology.

See Also Multilink Frame Relay (MFR)

MIB

Stands for Management Information Base, a database of information about a device that is managed using Simple Network Management Protocol (SNMP).

See Also Management Information Base (MIB)

microkernel

An operating system architecture in which the kernel is a small component with limited functionality that loads other components such as drivers and services into memory only as required to complete requested system tasks.

See Also kernel

Microsoft Certified Partner

An independent company that partners with Microsoft Corporation to provide IT (information technology) services to corporate, government, and small business customers.

Overview

The Microsoft Certified Partner Program is a partnership between Microsoft and independent companies where Microsoft provides IT services and products that enable these companies to provide solutions for their clients based on leading-edge Microsoft technologies. The program is global in scope and is the primary channel by which Microsoft solutions are provided. The Microsoft Certified Partner Program evolved primarily from the Microsoft Certified Solution Provider program formed in 1992, and it has over 31,000 partners worldwide.

Microsoft Certified Partners receive numerous benefits to help them deliver Microsoft solutions, including licenses for internal and marketing use of Microsoft products, access to prerelease software information and program codes, sales and marketing resources, technical training and support, and direct support from Microsoft staff. Partners participate in joint marketing ventures and receive valuable customer referrals.

The Microsoft Certified Partner Program has two levels of membership:

For More Information

Find out more about the Microsoft Certified Partner program at www.microsoft.com/certpartner.

Microsoft Certified Professional (MCP)

A family of technical certifications relating to proficiency in working with Microsoft products.

Overview

Microsoft Certified Professionals (MCPs) are individuals who have demonstrated in-depth knowledge of specific Microsoft operating system and application platforms. Different Microsoft certifications relate to different areas of expertise, knowledge, and skill. The various types of technical certifications offered by Microsoft Corporation can be summarized as follows:

A fourth area of Microsoft professional certification is the Microsoft Certified Trainer (MCT) designation, which is used to identify individuals qualified to deliver technical training in different Microsoft platforms and applications using Microsoft Official Curriculum (MOC) courseware at a Microsoft-authorized training center.

Employers can use Microsoft certifications for screening candidates for technical positions in their IT departments, and current employees can be encouraged to achieve this designation as part of their career advancement path. Certifications are achieved by passing exams administered by independent organizations such as Sylvan Prometric and Virtual University Enterprises, which together have more than 1400 testing centers worldwide.

Notes

Microsoft also has a nontechnical certification for users, including the Microsoft Office User Specialist (MOUS) certification.

For More Information

Find out about Microsoft certification at www.microsoft.com/trainingandservices.

See Also Microsoft Office User Specialist (MOUS) ,Microsoft Official Curriculum (MOC)

Microsoft Challenge Handshake Authentication Protocol (MS-CHAP)

An encrypted authentication scheme for Point-to-Point Protocol (PPP) sessions.

Overview

Microsoft Challenge Handshake Authentication Protocol (MS-CHAP) is Microsoft Corporation's version of the Internet standard Challenge Handshake Authentication Protocol (CHAP) defined in RFC 1994. Remote access servers use MS-CHAP to encrypt authentication sessions with remote access clients. Although CHAP requires the plaintext version of the remote user's password to be available on the access server for comparison purposes, MS-CHAP requires only the MD4 hash of the user's password to be available. The user's password information is thus stored in encrypted form on the access server, which makes MS-CHAP more secure than CHAP. MS-CHAP also differs from CHAP in that it employs a challenge/response packet format specifically designed for Windows platforms.

There are two different versions of MS-CHAP:

Implementation

Use of MS-CHAP is negotiated during the Link Control Protocol (LCP) portion of the PPP authentication process. Once the remote client requests authentication from the access server and negotiates MS-CHAP, the authentication process is encrypted in three steps:

  1. The access server sends the remote client a Challenge message consisting of a session ID and an arbitrary challenge string.

  2. The client sends the server a Response message consisting of the username (in plaintext), a hash of the challenge string, the session ID, and a one-way MD4 hash of the user's password.

  3. The server generates a duplicate version of the hash of the user's password and compares this with the Response message received from the client. If these match, the client is authenticated and the PPP session is initiated.

See Also Challenge Handshake Authentication Protocol (CHAP) ,Link Control Protocol (LCP) ,Point-to-Point Protocol (PPP) ,remote access

microsoft.com

The Web site for Microsoft Corporation, the worldwide leader in software for business and personal computing.

Overview

Microsoft's Web site is one of the largest and most complex in the world, as might be expected for a company having over 48,000 employees in over 60 regional offices worldwide. The microsoft.com site is also the main resource for those who use Microsoft products, including consumers, businesses, consultants, system integrators, and software developers. The staffs of IT (information technology) departments, sales and marketing, help desk, and business implementers can find useful information on microsoft.com for planning, designing, deploying, managing, maintaining, and troubleshooting solutions built using Microsoft products, platforms, and technologies.

Despite the size and complexity of the site, microsoft.com is actually amazingly simple to navigate! Nevertheless, finding information on a site as large as this is sometimes difficult, as many important Uniform Resource Locators (URLs) are not found on the site's home page. This article is designed to simplify this process of finding useful information for networking professionals by providing a summary of links to valuable information found on Microsoft's Web site. The URLs listed here are mostly top-level URLs that are unlikely to change frequently (many have remained the same for several years), and even if the site managers change them, these URLs are likely to be redirected automatically to the updated information. The URLs in the main table below are listed in short form as relative URLs; for example, /windows is short for www.microsoft.com/windows.

Note that a few URLs listed below are redirections to other Microsoft Web sites, such as the Microsoft Developer Network (MSDN) site at msdn.microsoft.com. And information about Microsoft's suite of developer products is actually found on the MSDN site rather than on the main www.microsoft.com site. For example, the site for Microsoft Visual Studio is found at msdn.microsoft.com/vstudio instead of www.microsoft.com/vstudio, but the latter URL actually redirects you to the former.

Essential Links for Networking Professionals on Microsoft.com

Relative URL

Information on . . .

/backofficeserver

Microsoft BackOffice Server home

/backstage

Inside view of microsoft.com operations

/billgates

Bill Gates's home page

/business

Microsoft business products and services

/catalog

Catalog of all Microsoft products

/certpartner

Microsoft Certified Partners home

/college

Microsoft jobs for college students

/ddk

Microsoft Windows Driver Development Kits

/directx

Microsoft DirectX home

/diversity

Microsoft Diversity home

/downloads

Microsoft download center (redirection)

/ebooks

Microsoft eBooks home

/education

Microsoft education home

/embedded

Windows Embedded home (redirection)

/enable

Microsoft accessibility information

/games

Microsoft games official Web site

/giving

Microsoft giving programs

/hcl

Hardware compatibility list (HCL) for Microsoft products

/hwdev

Windows driver and hardware development site

/insider

Microsoft consumer products and services

/jobs

Jobs at Microsoft

/kids

Free stuff and special offers for kids

/mac

Mactopia, Microsoft products for Macintosh

/mba

Microsoft jobs for MBAs

/mindshare

Mindshare User Group Support Program

/mobile

Mobile Devices and Pocket PC home

/mscorp

Corporate information about Microsoft

/msft

Microsoft investor relations

/museum

Microsoft museum online

/net

Microsoft .NET platform home

/office

Microsoft Office main site (see Note 1 below)

/partner

Microsoft for partners home

/piracy

Microsoft anti-piracy home

/presspass

Microsoft news and press releases

/reader

Microsoft Reader with ClearType

/security

Microsoft security home

/seminar

Microsoft Multimedia Central

/servers

Microsoft .NET Enterprise Servers (see Note 2 below)

/serviceproviders

Microsoft service providers home

/support

Product support home (redirection)

/technet

Microsoft TechNet home for IT professionals

/trainingand services

Microsoft training and certification home

/usability

Participate in product evaluations

/vstudio

Redirection to Microsoft Visual Studio .NET site on MSDN (see Note 3 below)

/windows

Microsoft Windows main site (see Note 4 below)

/winlogo

Microsoft Windows logo program

/worldwide

Microsoft worldwide information

The following table lists some other useful URLs in the microsoft.com domain; these, however, begin with a host prefix different from the usual www. prefix. Each of these sites represents a wealth of information useful to IT professionals.

Some Other Sites in the Microsoft.com Domain

URL

Description

communities.microsoft.com/ home/

Microsoft newsgroups, technical chats, and user groups

dgl.microsoft.com

Microsoft Office Design Gallery Live (redirection)

office.microsoft.com

Office Update home

mcsp.microsoft.com/home

Microsoft Certified Partner home

mcspreferral.microsoft.com

Find a Microsoft Certified Partner

msdn.microsoft.com

Microsoft Developer Network home

msdnisv.microsoft.com

Microsoft Business Connection for Partners

msevents.microsoft.com

Search for Microsoft events (redirection)

mspress.microsoft.com

Microsoft Press

murl.microsoft.com

Multi-University Research Laboratory

research.microsoft.com

Microsoft Research (MSR) home

search.microsoft.com

Search for info on microsoft.com (redirection)

shop.microsoft.com

Shop online at Microsoft

support.microsoft.com/ directory

Microsoft product support home

support.microsoft.com/search

Microsoft Knowledge Base (KB)

terraserver.microsoft.com

Microsoft Terraserver (redirection)

windowsupdate.microsoft.com

Windows Update home

Finally, here are a few other interesting and useful URLs to Web sites that are managed by or connected with Microsoft but are not part of the microsoft.com domain.

Some Microsoft Sites Outside the Microsoft.com Domain

URL

Description

bcentral.com

Microsoft small business resource center

expedia.com

Expedia.com home

hotmail.com

Hotmail

msn.com

MSN

msnbc.com

MSNBC home

passport.com

Microsoft Passport home (redirection)

webtv.com

MSN TV Service

windowslogo.com

Windows Logo Program (redirection)

windowsmedia.com

MSN site for music, radio, video, and more

Notes

  1. URLs for specific components of Microsoft Office are usually easy to guess. For example, to find the page for Microsoft Word, use www.microsoft.com/office/word, and to find the page for Microsoft Excel, use www.microsoft.com/office/excel. This does not always work, however; for instance, www.microsoft.com/frontpage for the Microsoft FrontPage home.

  2. Although the URL www.microsoft.com/servers provides a starting point for information about any of Microsoft's .NET Enterprise Servers, you can usually reach the page for a specific server by using the name of the server in the URL. For example, to find information on Microsoft Application Center 2000, use www.microsoft.com/applicationcenter; for information on Microsoft BizTalk Server 2000, use www.microsoft.com/biztalk; and so on. Sometimes the URL can be a little more cryptic, as in www.microsoft.com/isaserver for Microsoft Internet Security and Acceleration Server 2000. And sometimes the URL requires a specific page, such as www.microsoft.com/sharepoint/portalserver.asp, for Microsoft SharePoint Portal Server, although in this case there is a link from the default page www.microsoft.com/sharepoint to the main SharePoint Portal Server page.

  3. URLs for specific components of Microsoft Visual Studio are sometimes easy to guess. For example, the main page for Microsoft Visual Basic is msdn. microsoft.com/vbasic, but the page for Microsoft Visual C++ is msdn.microsoft.com/visualc instead. Do not forget to use msdn instead of www as the first portion of a URL for a developer product.

  4. URLs for versions of Microsoft Windows are usually obvious. For example, www.microsoft.com/windowsxp for Windows XP, www.microsoft.com/windows2000/ for Windows 2000, and so on. This pattern breaks, however, with Windows CE, where the URL is instead www.microsoft.com/windows/embedded/ce.

See Also Microsoft Corporation ,Microsoft Developer Network (MSDN) TechNet

Microsoft Consulting Services (MCS)

A service from Microsoft Corporation that provides direct assistance to large enterprises for planning, deploying, and supporting solutions based on Microsoft technologies.

Overview

Microsoft Consulting Services (MCS), in over 100 offices worldwide, has around 4000 Microsoft experts who can provide consulting services to help enterprises improve productivity, establish a competitive advantage, and make the most of their IT (information technology) investment. MCS offers a full range of programs, including enterprise application planning, distributed network architecture computing, e-commerce solutions, and more. Services offered by MCS include

For More Information

Visit www.microsoft.com/business/services/mcs.asp.

See Also Microsoft Operations Framework (MOF) ,Microsoft Solutions Framework (MSF)

Microsoft Corporation

The worldwide leader in software for business and personal computing.

Overview

Microsoft Corporation is organized around seven core groups focusing on different aspects of product services:

The current business leadership team for Microsoft includes

Microsoft is headquartered in a large campus located in Redmond, Washington. The main campus consists of 40 buildings that provide more than 4 million square feet (371,500 square meters) of office space. Microsoft employs more than 48,000 people worldwide, with almost half of these working in Washington State. Almost half of Microsoft's employees are involved in research and development, and most of the rest are in sales and marketing, with a small percentage in corporate operations and administration. The average age of a Microsoft employee is 34.3 years, and men outnumber women by almost three to one.

Microsoft operates subsidiary offices in more than 60 foreign countries and regions; more than half of Microsoft's revenue comes from these offices. Microsoft's net revenue for the fiscal year ending in June 2001 was $25.3 billion, with a net profit of $7.35 billion. Microsoft is traded on the Nasdaq using the symbol MSFT and has the largest capitalization of any corporation in the world.

History

September 2000 marked the 25th anniversary of Microsoft Corporation, and much has happened in those 25 years. In 1975, the personal computer (PC) was basically a toy for hobbyists, but today it represents the ubiquitous tool of business and consumers. Microsoft and its founder Bill Gates have played a major role in the evolution of the PC into what it has become. By striving to produce innovative products and services to meet customers' evolving needs, Microsoft seeks to realize its vision of "empowering people through great software-anytime, anyplace, and on any device."

Microsoft was founded as a partnership on April 4, 1975, by William H. Gates III and Paul G. Allen. Microsoft began its operations in Albuquerque, New Mexico, and earned only $16,005 in its first year of business. The company soon moved to the Seattle area, and by 1978 its sales had passed the $1 million mark. The company was incorporated on June 25, 1981, as Microsoft Corporation. Microsoft became a publicly traded company in 1986 at $21 per share.

In 1981, IBM decided to bundle Microsoft's MS-DOS operating system with the IBM PC. Microsoft continued to work with IBM throughout the 1980s to develop a successor to MS-DOS called OS/2 but abandoned this relationship in 1990. Meanwhile, Microsoft was developing its own graphical operating system called Microsoft Windows, the first release version of which appeared in 1985. Version 3 of Windows was released in 1990, and in 1991 Windows 3.1 became the standard desktop operating system for PCs everywhere. Since then, Microsoft has released a series of successors to Windows 3.1, including the popular Windows 95, Windows 98, and Windows Millennium Edition (Me) platforms for consumers and the Windows NT and Windows 2000 platforms for businesses. The latest versions of Windows, Windows XP and the Windows .NET Server family, offer state-of-the-art technology and performance for business and consumer computing platforms.

In 1989, Microsoft launched its first version of Office, a suite of desktop business productivity software applications. Office has evolved to its present version, Office XP, a powerful suite of tools for business workflow and productivity. Microsoft Office has become the leading office productivity suite used in the workplace today, just as Microsoft Windows has become the dominant desktop operating system.

In 1993, Microsoft released the first version of its 32-bit business computing platform, Windows NT. This operating system soon became the foundation for a suite of server business applications developed by Microsoft and integrated into a single platform called Microsoft BackOffice. The BackOffice suite of server applications included Microsoft Exchange, which has become the leading enterprise messaging and collaboration platform, and Microsoft SQL Server, which is on its way to becoming the leading database platform for business knowledge management solutions. In 2000 a new vision was articulated by Microsoft for the enterprise to succeed BackOffice. This new vision of software as services is based on Microsoft's .NET family of server products and is the major focus of Microsoft research and development at the start of the new millennium.

Notes

Corporate contact information for Microsoft is

Microsoft Corporation One Microsoft Way Redmond, WA 98052-6399 USA Tel: (425) 882-8080 E-fax: (425) 706-7329

For More Information

To find out more about Microsoft and Microsoft products, visit www.microsoft.com

See Also BackOffice , .NET platform

Microsoft Data Access Components (MDAC)

A set of Microsoft technologies that provide access to information stored in a broad range of data sources.

Overview

Microsoft Data Access Components (MDAC) include the following:

Developers can use MDAC components for building distributed Web applications and e-commerce solutions.

See Also ActiveX Data Objects (ADO) ,OLE DB ,open database connectivity (ODBC)

Microsoft Developer Network (MSDN)

Microsoft Corporation's portal and support program for its developer community.

Overview

Microsoft Developer Network (MSDN) provides developers who work with Microsoft programming platforms with tools, information, training, and events to help them develop applications for Microsoft Windows platforms. The core of the MSDN program is the MSDN Web site, which provides developers with

MSDN also offers paid subscriptions of various levels, providing developers through monthly CD shipments with news, product documentation, code samples, technical articles, and software for various Microsoft platforms, applications, and programming tools. Developers can also receive regular updates of information on the MSDN site by subscribing to MSDN Flash, a semimonthly e-mail newsletter with news, event listings, and important announcements.

For More Information

Visit MSDN online at msdn.microsoft.com

See Also microsoft.com ,TechNet

Microsoft Disk Operating System (MS-DOS)

An operating system created by Microsoft Corporation in 1981 for the first IBM personal computer (PC).

Overview

Microsoft Disk Operating System (usually known only by its acronym, MS-DOS) was a 16-bit operating system that used a text-based command-line interface (CLI) for executing commands. The DOS part of MS- DOS indicates that the operating system is disk- based-that is, the operating code resides on a disk (initially a floppy disk and later the hard disk). MS- DOS revolutionized the nascent PC market by providing users with low-level access to operating system functions such as managing disk storage and memory resources, creating directories and manipulating files, and creating batch files for automating tasks. MS-DOS was so popular that other vendors soon produced their own versions of DOS, including IBM's PC-DOS and Novell's DR-DOS.

History

MS-DOS evolved rapidly and went through a number of revisions that added greater functionality and power to the basic operating system. Its final version, MS- DOS 6.22, was so stable and reliable that even though it had already been superseded for several years by Microsoft Windows 3.1, Microsoft's popular graphical operating system, in the late 1990s some companies could still be found running MS-DOS-some examples being conservative companies such as banks and other financial institutions.

Few companies still run MS-DOS today, however, as Microsoft Windows 2000 provides similar levels of stability and reliability, plus the ease of use of a graphical user interface (GUI) coupled with the processing power and storage capacity of modern Intel-based hardware platforms. Computer technicians still occasionally use MS-DOS, however, mainly in the form of MS-DOS boot disks for repairing problem systems.

History of MS-DOS Versions

Version

Release Date

Features

1

August 1981

Designed for the IBM PC, the initial version was distributed on a single 160-kilobyte (KB) floppy disk and ran in 8 KB of RAM.

1.1

May 1982

Support was added for double- sided 320-KB disks.

2

March 1983

Designed for the IBM PC/XT, this version added support for hard disks, hierarchical directories, background printing, and third-party device drivers.

2.1

October 1983

Designed for the short-lived IBM PCjr.

3

August 1984

Designed for the new IBM PC/AT, this version added support for 1.2-megabyte (MB) floppy disks and hard drives larger than 10 MB.

3.1

March 1985

Added support for networking and file sharing.

3.2

January 1986

Added support for the new 3.5-inch floppy disks.

3.3

April 1987

Designed for the IBM PS/2, this version added commands and support for international versions.

4.01

February 1988

This version added the Mem command, the MS-DOS shell, and support for hard drives larger than 32 MB.

5

May 1991

For this version, the MS-DOS shell was redesigned, task- swapping was added, and more extensive help, undelete, unformat, and memory- management tools were added.

6

March 1993

Added the MemMaker utility, real-time disk compression, multiple boot configurations, and antivirus and backup utilities.

6.2

October 1993

Added Scandisk for low-level disk checking.

6.22

February 1994

Added DriveSpace disk compression.

Architecture

MS-DOS commands come in two types:

The core of the MS-DOS operating system consists of three primary files in the root of the system partition:

In addition, two other text files help control the MS-DOS boot process:

The remaining external MS-DOS commands and utilities are by default found in the directory C:\DOS.

Notes

You can add networking functionality to MS-DOS by using the Microsoft Network Client 3 for MS-DOS add-on. You can create installation disks for this software using the Network Client Administrator tool in Microsoft Windows NT.

See Also Microsoft Windows

Microsoft Foundation Classes (MFC)

A set of object-oriented interfaces to the Microsoft Windows application programming interface (API).

Overview

Microsoft Foundation Classes (MFC) include classes, global functions, global variables, and macros that provide a framework for developing applications for Windows platforms. MFC encapsulates much of the Windows API by providing classes that represent key Windows objects, such as windows, controls, dialog boxes, brushes, and fonts. Programmers can develop Windows-based applications by using a combination of C++ code and MFC instead of calling Windows API functions directly. Many MFC class member functions actually call the encapsulated Windows API functions. Note that you cannot call MFC class member functions directly-you have to instantiate the class first.

Architecture

The majority of the MFC are derived by inheritance from the root class named CObject. CObject provides support for serializing data and obtaining run-time class information, although you do not need to derive new classes from CObject if you do not need these capabilities. Other classes in MFC include the following:

See Also ActiveX ,application programming interface (API) ,Data Access Objects (DAO) ,open database connectivity (ODBC)

Microsoft Management Console (MMC)

A tool and software framework for administering systems running Microsoft Windows 2000 or later.

Overview

The Microsoft Management Console (MMC) was first included with Microsoft Windows NT Option Pack to provide an integrated management framework for a wide range of administrative tasks. The MMC later became the standard administrative interface for managing the Microsoft Windows 2000 operating system and server applications such as Microsoft Exchange Server and Microsoft SQL Server.

MMC offers the following features:

Windows 2000 and later versions include a number of preconfigured consoles known as administrative tools. These tools are used for performing specific administrative tasks, and you can access them using the Administrative Tools shortcut in the Programs group of the Start menu.

Implementation

By itself, the MMC does not provide any system or network management capability. Instead, it provides an environment in which administrative tools called snap- ins can be run. A snap-in is a software component that provides some system or network management capability for administrators to perform standard tasks.

MMC console. Computer Management, an example of an MMC console.

When one or more snap-ins have been added to a blank console, the configuration can be saved as an .msc file and then shared with other administrators by e-mail or by sharing it on a file server. The MMC also supports deregulation of different levels of management capability to other administrators. For example, an administrator might

The MMC user interface presents a hierarchical view of your network resources in a two-pane view similar to that of Windows Explorer. The left pane (the Scope pane) shows a hierarchical view of the administrative namespace of manageable nodes (network objects), while the right pane (Results pane) shows the contents, services, or configuration items of the selected node in the Scope pane. You perform management tasks by using the menu or toolbars, opening property sheets for nodes, and accessing Web pages in the Results pane.

Microsoft Market

A business-to-business (B2B) procurement system created and used internally by Microsoft Corporation.

Overview

Microsoft Market is an Internet-based procurement system developed by Microsoft to streamline operations with its supply-chain partners. Although Microsoft Market is used internally by Microsoft and is not open to the public, it is an excellent example of how B2B systems using the Internet can reduce the cost of doing business with a company's business partners.

Before deploying Microsoft Market, Microsoft's procurement costs typically were $60 per transaction. With Microsoft Market implemented, this overhead has dropped to less than $5 per transaction. Using Microsoft Market, purchases of goods and services from partners and vendors can be performed using a standard Web browser interface. Microsoft Market has helped Microsoft lower its procurement costs and speed up its business cycle; it has also reduced sales and marketing costs and has created new sales opportunities and improved customer service. Microsoft Market puts Microsoft in the forefront of companies implementing B2B Internet technologies in streamlining business process to gain a competitive advantage in the marketplace.

See Also B2B ,e-business

Microsoft Office User Specialist (MOUS)

A Microsoft certification for users of Microsoft Office.

Overview

The Microsoft Office User Specialist (MOUS) certification enables individuals to demonstrate their in-depth knowledge of the Office suite of business productivity tools. The MOUS certification can be used for skills assessment to help them find the qualified knowledge workers they need for business success.

Depending on the version of Office considered, the MOUS program provides different levels of certification to help identify proficient, expert, and master levels of competency in using the product. MOUS certification is also available for Microsoft Project 2000.

For More Information

Find out more about MOUS at www.microsoft.com/trainingandservices.

See Also Microsoft Certified Professional (MCP)

Microsoft Official Curriculum (MOC)

Courseware developed by Microsoft Corporation for training in Microsoft products and technologies.

Overview

Microsoft Official Curriculum (MOC) courseware is designed to provide comprehensive training in Microsoft products to IT (information technology) professionals who develop, implement, administer, and support business solutions based on Microsoft platforms, applications, and tools. MOC courseware are available for the full range of Microsoft products, including all versions of Microsoft Windows, the .NET Enterprise Server family, and Microsoft programming platforms. MOCs are available in different forms, including

MOCs are the standard means for IT professionals to prepare for exams leading to Microsoft Certified Professional (MCP) certifications.

For More Information

Find out more about MOCs at www.microsoft.com/trainingandservices.

See Also Certified Technical Education Center (CTEC) ,

Microsoft Operations Framework (MOF)

A service by Microsoft Corporation that provides technical guidance on Microsoft technologies to large enterprises.

Overview

Microsoft Operations Framework (MOF) is part of the Microsoft Consulting Services (MCS) program and is designed to help enterprises achieve mission-critical reliability, availability, supportability, and manageability in their use of Microsoft products and technologies. MOF provides various tools to enterprises, including assessment tools, best practices, case studies, operations guides, support tools, templates, and white papers. Issues covered in these resources address the people, process, technology, and management issues that arise in large, heterogeneous IT (information technology) environments. MOF also provides structured support programs and other services that can enhance the capabilities of your in-house IT operations.

For More Information

Visit www.microsoft.com/business/services/ mcsmof.asp.

See Also Microsoft Consulting Services (MCS)

Microsoft RAS Protocol

A legacy remote access protocol from Microsoft Corporation.

Overview

Microsoft RAS Protocol is a remote access protocol developed for Microsoft Windows for Workgroups 3.11 and Windows NT 3.1 as an alternative to Serial Line Internet Protocol (SLIP) that was in common use at that time on the UNIX platform. Microsoft RAS Protocol enables legacy Windows for Workgroups and Windows NT clients to dial in and connect to a remote access server (RAS) using the NetBIOS Enhanced User Interface (NetBEUI) protocol. Once the remote client establishes a connection, the remote access server acts as a NetBIOS gateway to enable the client to access other servers on the network using NetBEUI, NetBIOS over Internetwork Packet Exchange (IPX), or NetBIOS over Transmission Control Protocol/Internet Protocol (TCP/IP).

Microsoft RAS protocol is supported by later versions of Windows only for purposes of backward compatibility. You need RAS Protocol only if you are connecting to a Microsoft Windows remote access server from legacy Windows clients running Windows for Workgroups 3.11 or Windows NT 3.1. Apart from this usage, Microsoft RAS Protocol has been entirely superseded by the Point-to-Point Protocol (PPP), the industry-standard protocol for remote access.

See Also NetBEUI ,NetBIOS ,Point-to-Point Protocol (PPP) ,Serial Line Internet Protocol (SLIP) ,UNIX

Microsoft Research (MSR)

A branch of Microsoft Corporation concerned with advanced research in computer science and technologies.

Overview

Microsoft Research (MSR) pursues research in a variety of different technologies that are viewed as having the potential to change the face of computing over the next decade. These technologies include artificial intelligence (AI), computer vision, speech processing, quantum computing, and advances in operating systems and programming language and tools that will make computers friendlier and easier to use.

MSR was founded in 1991 and was the first research laboratory for basic computer science established by any software company. MSR currently has over 600 researchers operating in four different laboratories around the world:

For More Information

Visit MSR at www.research.microsoft.com

Microsoft Solutions Framework (MSF)

A Microsoft Corporation training program for large companies that provides guidance on planning, building, and deploying solutions based on Microsoft products and technologies.

Overview

Microsoft Solutions Framework (MSF) addresses training in the areas of enterprise application development, architecture design, component design, and deployment of infrastructure. MSF training is based on best practices from real-life consulting done by Microsoft Consulting Services (MCS), Microsoft Certified Partners, and others who develop Microsoft solutions for business needs. The MSF program is offered to enterprises through MCS.

MSF is based on a series of guidelines-based models, including

For More Information

Visit www.microsoft.com/business/services/mcsmsf.asp.

See Also Microsoft Certified Partner ,Microsoft Consulting Services (MCS)

Microsoft Technology Center (MTC)

A state-of-the-art training center for learning about Microsoft technologies and solutions.

Overview

Microsoft Technology Centers (MTCs) are laboratories where enterprise IT (information technology) staff can work side by side with Microsoft experts and partners to design, develop, and test solutions based on Microsoft applications and platforms. MTCs are located in various cities across North America and offer a broad range of facilities, programs, and expertise for developing technology solutions for specific business needs. MTCs also offer boot camps to provide environments for rapid learning of state-of-the-art Microsoft technologies and solutions such as BizTalk and the .NET framework. MTCs are available to enterprises through Microsoft Consulting Services (MCS).

For More Information

Visit www.microsoft.com/business/services/mtc.asp.

See Also Microsoft Consulting Services (MCS)

Microsoft Windows

Microsoft Corporation's flagship operating system platform.

Overview

Microsoft Windows is a family of operating systems that lie at the core of Microsoft's strategy to make PCs easier to use, reduce the cost of PC ownership, advance the PC platform in the enterprise, and integrate PCs with the Internet.

Windows operating systems evolved in the 1990s from the earlier MS-DOS, Microsoft's powerful text-based disk operating system for personal computers. The first versions of Windows were 16-bit operating systems that essentially ran on top of MS-DOS and provided an easy-to-use graphical user interface (GUI) that made computers easier to learn and use. These 16-bit versions included Windows 3.1 and Windows for Workgroups 3.11, the network-aware version of Window 3.1. In the middle of the decade, Microsoft released a new version of Windows called Windows 95, a desktop operating system that included a new and more powerful GUI, support for preemptive multitasking, and enhanced management of system resources. Windows 95 evolved into Windows 98 and then Windows Millennium Edition (Windows Me).

Meanwhile, Microsoft also developed a fully 32-bit operating system called Windows NT. This new operating system was more powerful, robust, and secure than the Windows 95/98/Me family and was intended for business users instead of consumers. Although Windows NT 3.51 was based on the Windows 3.1 GUI, the popular Windows NT 4 employed a GUI similar to Windows 95. Windows NT 4 was offered in a variety of flavors targeted at the desktop, local area network (LAN) server, and enterprise server operating system markets.

As the decade concluded, Windows NT was superseded by the powerful Windows 2000 family of operating systems, which offered great stability and powerful features foundational for building strategic e-business solutions for the enterprise. Most recently, two new Windows platforms have evolved:

A separate evolutionary path for Windows has produced the Windows CE 3 operating system, a lightweight version of Windows designed for handheld PCs and appliances. Windows NT Embedded, another flavor of Windows, is a low-footprint version designed for industrial control systems and other devices with limited processing and memory.

The table below lists the current versions of Windows. For more information about particular versions, see the corresponding entries in Chapter W of this book.

Current Microsoft Windows Operating Systems

Version

Description

Windows .NET Server

Comes in Standard, Enterprise, Datacenter, and Web Server versions and is the successor to the Windows 2000 family of business operating systems

Windows XP

Comes in Professional and Home Edition and is the successor to both the Windows Me and Windows 2000 Professional lines of business and consumer Windows desktop operating systems

Windows 2000

Comes in Professional, Server, Advanced Server, and Datacenter Server versions and is the successor to the Windows NT 4 family of business operating systems

Windows Me

Successor to Windows 95 and Windows 98 consumer Windows desktop operating systems

Windows CE

Lightweight version of Windows designed for handheld computing devices and computer appliances

Windows NT Embedded

Small-footprint version of Windows NT designed for industrial control systems and small devices

For More Information

Visit the Windows home at www.microsoft.com/windows.

See Also Windows 3.1 ,Windows 95 ,Windows 98 ,Windows 2000 ,Windows CE ,Windows Me (Windows Millennium Edition) ,Windows .NET Server ,Windows NT ,Windows XP

middleware

Application logic between the client and data sources.

Overview

The concept of middleware emerged in the mid-1990s as a solution to scalability problems inherent in the traditional two-tier client/server computing paradigm. In a typical business computing system, client machines access information stored on back-end database servers for accessing inventory, sales, invoicing, and other business information. The scalability issue arises as databases grow in size and complexity-then the problem becomes where to implement the application logic for best performance. Choosing a "fat client" approach moves more of the processing load to the client, but the result is that whenever the business application is upgraded or patched, new software needs to be installed on every client across the enterprise. Another approach is to move application logic to the database servers by using centrally stored procedures, but this increases the processing load on the servers considerably.

A third approach is to use middleware, which represents a third tier of software stored on a special server separate from the client and the database server. Employing middleware means that thin clients can be used instead of fat ones, that desktop systems do not need to be updated as frequently, and that some of the processing load is siphoned off from the database servers. When application logic needs to be updated, it can be done quickly and easily on the servers running the middleware. The result of proper implementation of middleware in the enterprise is greater scalability and better performance of line-of-business applications that use database technologies, such as Enterprise Resource Planning (ERP) and Customer Relations Management (CRM) applications.

Middleware technologies are also used in other scenarios. For example, cellular systems can allow mobile users to access company intranets using personal digital assistants (PDAs) and other devices. This is generally done by implementing a middleware server that connects the mobile devices to the database servers. An example of such a middleware application is Microsoft Mobile Information Server, one of the new .NET Enterprise Servers from Microsoft Corporation. Another type of middleware system is a wireless e-mail gateway that connects mobile users with handheld devices such as Research In Motion's Blackberry with Internet mail hosts. Such middleware servers are often available from vendors as either installable applications or as self-hosted (stand-alone) appliances.

As enterprise systems grow more complex, three-tiered systems often grow into n -tiered systems that employ several levels of middleware. The principles are still the same, though: improving performance and scalability by isolating application logic from both the client and the database server.

Notes

The term middleware is sometimes used in other contexts. For example, it can refer to the various technologies and protocols used to connect the clients with the back-end servers. These technologies can include remote procedure calls (RPCs), synchronous transactions, asynchronous message passing mechanisms, and distributed object oriented programming technologies such as Microsoft's Distributed Component Object Model (DCOM), Sun Microsystems' JavaBeans, and The OpenGroup's Common Object Request Broker Architecture (CORBA).

See Also client/server , Common Object Request Broker Architecture (CORBA) ,Customer Relationship Management (CRM) ,Distributed Component Object Model (DCOM) ,enterprise resource planning (ERP) ,Java , remote procedure call (RPC)

MIF

Stands for Management Information Format (MIF), a standard format for describing hardware and software management information.

See Also Management Information Format (MIF)

Millennium

An advanced distributed computing architecture being developed by Microsoft Research.

Overview

Millennium is a radically new vision for distributed computing in which the automatic distribution of application processing is the norm rather than the exception. The Millennium paradigm assumes that all applications are intrinsically distributed across the network rather than locally centralized on servers. Millennium is based on an advanced network-aware operating system than can dynamically distribute processing tasks and manage resources across the network. Millennium is based on four key concepts:

One of the Millennium prototype projects is called Coign, which supports the self-configuration and self-tuning of Component Object Model (COM) objects across a network. Coign supports the dynamic partitioning of applications to improve performance and simplify the management of distributed applications.

For More Information

Find out more at www.research.microsoft.com/sn/ millennium.

See Also Component Object Model (COM) , Jini ,

MIME

Stands for Multipurpose Internet Mail Extensions, extensions to Simple Mail Transport Protocol (SMTP) that allow multipart and binary messages to be sent using e-mail.

See Also Multipurpose Internet Mail Extensions (MIME)

mirrored volume

A type of fault tolerance supported by Microsoft Windows 2000 and Windows .NET Server.

Overview

In Microsoft Windows 2000 and Windows .NET Server, a mirrored volume is a type of volume created with Disk Management that duplicates data on two separate physical disks. This provides a measure of fault tolerance, for if one physical disk fails, the data is still accessible from the second disk.

In Windows 2000 and Windows .NET Server, mirrored volumes cannot be extended and they must be created on dynamic disks. Although mirrored volumes perform read operations more slowly than RAID 5 volumes, they do execute write operations more quickly.

Notes

The Windows NT equivalent of a mirrored volume was called a mirror set.

See Also dynamic volume , redundant array of independent disks (RAID)

mirroring

Maintaining an identical copy of an information store or disk system for fault tolerance purposes.

Overview

Mirroring is one of several different strategies for integrating fault tolerance into storage systems. Using mirroring, two identical copies of mission critical data are maintained, and if the primary system goes down, the secondary system can be brought online quickly with little or no loss of data. Mirroring can be implemented at various levels from disk mirroring (two identical disk drives within a server) to server mirroring (two identical servers having the same data and running at different locations) to mirroring of enterprise storage solutions such as storage area networks (SANs). Mirroring at higher levels is commonly used in financial institutions such as banks and brokerages to ensure speedy recovery in the event of a disaster.

Mirroring is not a replacement for tape backup systems; instead, it complements the standard backup cycle by providing additional redundancy and fault tolerance. Mirroring of enterprise servers is a popular solution in high availability environments because the recovery window (the time to bring a mirrored system online after a disaster) is typically less than an hour, while the time to restore a system completely from backup sets may be 24 to 48 hours. Mirroring does have the disadvantage of being expensive compared to tape backup, since duplicate storage systems are required to implement it.

Implementation

Mirroring can implemented using several techniques:

See Also backup , electronic tape vaulting ,fault tolerance ,frame relay ,Integrated Services Digital Network (ISDN) , storage, storage area network (SAN), storage service provider (SSP), wide area network (WAN)

mirror set

A type of fault tolerance supported by Microsoft Windows NT.

Overview

Mirroring is the process whereby data is written simultaneously to two disks, making one disk drive an exact mirror image of the other. This form of fault tolerance technology is properly known as RAID-1 but is called a mirror set in Windows NT (Windows 2000 calls this same configuration a mirrored volume).

The duplication of disk information in a mirror set provides the system with data redundancy, for if one half of the mirror set fails, the system can continue to operate using the duplicate of the failed disk. In Windows NT, mirror sets are commonly used to provide fault tolerance for the system and boot partitions. You can use Disk Administrator, the administrative tool in Windows NT, to create, break, and reestablish mirror sets. In Windows 2000, the corresponding administrative tool is called Disk Management.

For read operations, mirror sets are slower than other redundant array of independent disks (RAID) configurations such as stripe sets with parity, but on the other hand they are faster for write operations.

See Also fault tolerance , redundant array of independent disks (RAID)

mixed mode

A mode of running Microsoft Windows 2000 domain controllers.

Overview

When Windows 2000 domain controllers are running in mixed mode, they are backward compatible with domain controllers running the earlier Windows NT operating system. This is because Windows 2000 domain controllers running in mixed mode use Windows NT LAN Manager (NTLM) as their authentication protocol instead of Kerberos, which allows Active Directory directory service to communicate with downlevel Windows NT domain controllers.

Mixed mode has the following special considerations:

Implementation

Mixed mode is designed to be used mainly as a temporary solution during the process of migrating a Windows NT-based network to a Windows 2000-based one. Once the migration is complete, you can switch your Windows 2000 domain controllers from mixed mode to native mode. Note that this is a process that can be performed only once-you cannot change native mode domain controllers back to mixed mode.

Windows 2000 domain controllers are installed in mixed mode by default when you run the Active Directory Installation Wizard to upgrade a member server to a domain controller. To change domain controllers from mixed mode to native mode, use the Active Directory Users and Computers console. During a migration from Windows NT to Windows 2000, after you have upgraded all your downlevel domain controllers to Windows 2000 you should change your domain to native mode in order to take advantage of Windows 2000's special features, such as multimaster replication, nesting of groups, and universal groups. If you are deploying a pure Windows 2000-based network, however, you should configure your domain controllers to run in native mode from the start.

See Also Active Directory ,Active Directory Users and Computers ,domain controller ,domain modes ,Kerberos ,native mode ,universal group ,Windows NT LAN Manager Authentication

MLP

Stands for Multilink Point-to-Point Protocol, a wide area network (WAN) protocol for aggregating multiple Point-to-Point Protocol (PPP) connections.

See Also Multilink Point-to-Point Protocol (MPPP)

MLS

Stands for Multilayer Switching, a LAN switching term that can have several meanings.

See Also Multilayer Switching (MLS)

MMC

Stands for Microsoft Management Console, a tool and software framework for administering systems running Microsoft Windows 2000 or later.

See Also Microsoft Management Console (MMC)

MMDS

Stands for Multipoint Multichannel Distribution Service, a fixed wireless broadband networking architecture.

See Also Multipoint Multichannel Distribution Service (MMDS)

M-node

A NetBIOS name resolution method used by Microsoft Windows NT.

Overview

Name resolution is the process of converting the name of a host on the network into a network address (such as an Internet Protocol [IP] address). Name resolution must be performed in order to establish communication over a Windows NT network. M-node is one of four basic methods supported by Windows NT for resolving NetBIOS host names-that is, computer names-into IP addresses.

If a computer running Windows NT is configured as an M-node machine, it first tries to use broadcasts to resolve the names of the hosts, similar to a B-node machine. If name resolution fails this way (for example, if broadcasts are stopped by routers from reaching computers on other subnets), Windows NT tries to use a NetBIOS name server to resolve names of other hosts on the network, similar to a P-node machine. A Windows NT server with the Windows Internet Naming Service (WINS) configured on it is a typical example of a NetBIOS name server. The M in the term M-node stands for mixed , as M-node is a mixture of B-node and P-node, in that order.

Notes

M-node is defined in RFCs 1001 and 1002.

See Also B-node ,H-node ,NetBIOS name resolution ,P-node

Mobile Broadband System (MBS)

A proposed fourth-generation (4G) mobile cellular communication system.

Overview

The Mobile Broadband System (MBS) is a project of the Research and Technology Development in Advanced Communications Technologies in Europe (RACE) program. MBS is designed to extend mobile communications to support broadband communication at multimegabit data transmission rates of up to 155 megabits per second (Mbps). MBS is seen as a key enabler for transmission of multimedia video and audio content on cellular communications systems.

MBS is basically a wireless extension to broadband ISDN (B-ISDN), a technology that has been talked about for many years and is still in the developmental stage. The underlying transport used in MBS is Asynchronous Transfer Mode (ATM), and the proposed MBS standard from the International Telecommunication Union (ITU) is intended to fully integrate MBS with the ATM-based Integrated Broadband Communications Network (IBCN), also proposed for Europe.

See Also Asynchronous Transfer Mode (ATM) ,broadband ISDN (B-ISDN) ,cellular communications ,International Telecommunication Union (ITU)

Mobile Execution Environment (MExE)

An emerging standard for running Java applications on mobile phones.

Overview

Mobile Execution Environment (MExE) is a proposal of the European Telecommunications Standards Institute (ETSI) to enable mobile devices with limited processing power such as cell phones and personal digital assistants (PDAs) to run Java applications. Using MExE, for example, a mobile knowledge worker could easily access enterprise applications using his cell phone.

MExE is designed to reduce the footprint required for running Java on small mobile devices. This is done by partly compiling client-side Java applications on middleware servers, thus reducing the processing and memory requirements needed to run a full Java interpreter on these mobile devices. Wireless Application Protocol (WAP) is then used as a transport for delivery of information to the mobile devices.

See Also cellular communications ,Java ,Wireless Application Protocol (WAP)

Mobile Information Server

Microsoft Corporation's .NET Enterprise Server for enabling wireless access to enterprise data.

Overview

Microsoft Mobile Information 2001 Server provides enterprises with a platform for allowing users to access data stored on corporate intranets and database servers from mobile devices such as cell phones and wireless personal digital assistants (PDAs). Mobile Information Server thus extends the reach of .NET enterprise applications by bringing content to the mobile knowledge worker. Using Mobile Information Server, people can have access to corporate e-mail, calendars, contacts lists, tasks, and any other desired information.

Mobile Information Server includes two components:

Mobile Information Server offers an extensible architecture that supports connectors for allowing mobile users to access information stored on any type of data server or information store. Mobile Information Server supports end-to-end security, load balancing and clustering, and content replication and delivery.

For More Information

Find out more at www.microsoft.com/miserver.

See Also cellular communications , Exchange Server , .NET Enterprise Servers, SQL Server

Mobile Services Initiative (M-Services)

A proposed set of standards for broadband cellular Internet access.

Overview

Mobile Services Initiative (M-Services) is designed to speed the deployment of broadband cellular Internet access. The aim is to bring standards and consistency by providing guidelines for graphical displays on cellular handsets, transmission schemes for accessing music and video, and other functions. M-Services is designed to overcome some of the problems inherent in Wireless Application Protocol (WAP), the first generation of cellular Internet access technologies. WAP is generally viewed as too slow and complex for mobile platforms, and application development for this platform has tended to result in a variety of proprietary extensions.

M-Services focuses on providing guidelines for application development on General Packet Radio Services (GPRS), a 2.5G upgrade to existing second-generation (2G) Global System for Mobile Communications (GSM) cellular systems used widely in Europe and becoming more popular in North America. GPRS is viewed as a stepping stone to true third-generation (3G) broadband mobile communications systems, which are now not likely to be widely deployed until 2005.

M-Services is supported by the GSM Association, by manufacturers of mobile technologies such as Ericsson, Motorola, and Nokia, and by mobile operators such as France Telecom and Telecom Italia Mobile.

See Also 2.5G ,3G ,cellular communications ,General Packet Radio Service (GPRS) ,Global System for Mobile Communications (GSM) ,Wireless Application Protocol (WAP)

Mobile Transport Serving Office (MTSO)

A component of a cellular communications system.

Overview

In a cellular communications system, the region where coverage is desired is broken up into a series of small overlapping areas called cells. Each cell is serviced by a base station, which has a transceiver (transmitter/receiver) for sending and receiving calls with mobile users in that area. For communications to take place, however, frequencies or time slots (depending on the cellular technology used) must be allocated without conflict to users in different cells, a function performed by a central station called a Mobile Transport Serving Office (MTSO). The MTSO thus coordinates the activities of the different transceivers within a given region and provides a land-based (wireline) link between these transceivers and the telco central office (CO) for connection to the Public Switched Telephone Network (PSTN). Connections between base stations and the MTSO, and between the CO and the MTSO, are usually made using telco trunk lines.

Mobile Transport Serving Office (MTSO). How the MTSO connects a cellular communications system to the Public Switched Telephone Network (PSTN).

See Also cellular communications ,central office (CO) ,Public Switched Telephone Network (PSTN) ,telco

MOC

Stands for Microsoft Official Curriculum, courseware developed by Microsoft Corporation for training in Microsoft products and technologies.

See Also Microsoft Official Curriculum (MOC)

modem

A device that enables digital data transmission to be transmitted over telecommunications lines.

Overview

The term modem , which stands for modulator/demodulator, generally refers to analog modems, data communications equipment (DCE) that converts the digital signals used by data terminal equipment (DTE) such as computers into analog sound waves that can be transmitted over the analog local loop portion of the Public Switched Telephone Network (PSTN). Analog modems generally have two interfaces: an RS-232 serial transmission interface for connecting to the DTE, which is usually a computer; and an RJ-11 telephone interface for connecting to a standard four-wire PSTN telephone jack.

Analog modems are popular and come in various types:

Another kind of modem is the digital modem, one example of which is the Integrated Services Digital Network (ISDN) terminal adapter. Although a digital modem does not convert digital signals into analog and vice versa, a form of modulation called line coding does take place. Line coding is used to modulate the digital signals from the DTE, which is typically a switch or router connecting to a local area network (LAN), into digital signals that can be transmitted over the specially conditioned ISDN telephone. Other popular types of digital modems include cable modems and Digital Subscriber Line (DSL) modems. Digital modems use a variety of serial interfaces, including RS-232, RS-449, V.35, X.21, and High Speed Serial Interface (HSSI).

See Also analog modem ,cable modem ,data communications equipment (DCE) ,data terminal equipment (DTE) ,digital modem ,Digital Subscriber Line (DSL) ,Integrated Services Digital Network (ISDN) ,ISDN terminal adapter ,Public Switched Telephone Network (PSTN) ,RS-232 ,serial transmission

modem eliminator

A device for connecting two pieces of data terminal equipment (DTE).

Overview

Modem eliminators provide an easy way to connect two pieces of DTE without a modem. Modem eliminators accomplish this by simulating a synchronous transmission data link through generating timing and handshaking signals. Modem eliminators can use a variety of serial interfaces, including RS-232, RS-422/485, RS-530, V.25, or X.21, and can operate at speeds of up to 2.048 megabits per second (Mbps).

Modem eliminator. Using a modem eliminator to connect two DTE over a serial link.

Modem eliminators are similar to line drivers, but modem eliminators simply connect two pieces of DTE and line drivers connect DTE to data communications equipment (DCE). Also, line drivers regenerate signals and thus extend the maximum possible distance, but modem eliminators do not. Finally, line drivers are used in pairs (one at each end of the line), and modem eliminators are used singly and are placed at the midpoint between the two devices being connected.

See Also data communications equipment (DCE) , data terminal equipment (DTE) ,line driver , serial transmission

modem sharing

Any method for sharing a modem among several computers.

Overview

Modem sharing is an inexpensive method for small networks to share a single Internet or remote access connection. For example, in a Small Office/Home Office (SOHO) environment, a cable modem or Digital Subscriber Line (DSL) modem can be used to share a single high-speed Internet connection among several users. Another scenario is in remote branch offices where a single shared modem can be used for occasional remote access connectivity over a wide area network (WAN) link to the head office.

Modem sharing can be implemented in two ways:

Marketplace

Modem sharing hardware for DSL and cable Internet services are becoming popular. Some examples of broadband routers from different vendors include the ZyXEL Prestige 310, Netopia 9100, and WatchGuard SOHO.

Notes

If you use modem sharing to provide your network with high-speed connectivity to the Internet, be sure to use a firewall to secure your network against outside intrusion. If you use Windows 2000, another good step is to turn off Internet File Sharing.

See Also broadband Internet access , cable modem ,Digital Subscriber Line (DSL) ,hub ,Integrated Services Digital Network (ISDN) , router

modulation

Converting analog signals into digital signals (the reverse process is called demodulation).

Overview

Modulation is the process by which digital information is encoded into analog electrical signals for transmission over a medium. Digital information is usually binary information, as represented by a series of 1s and 0s, and must be converted into analog electrical signals (voltages) for transmission over wires (or into light waves for transmission over fiber-optic cabling).

Modulation is often confused with signal encoding or line coding, as these processes are inherently similar. However, for transmission using modems over the Public Switched Telephone Network (PSTN) lines, the process is generally called modulation, but for transmission over digital lines such as T1, frame relay, and Integrated Services Digital Network (ISDN) lines, the term encoding or line coding is usually used instead. Some reference works tend to blur this distinction, however, and there is usually some overlap between the concepts.

Types

Modulation forms the basis of the digital-to-analog converter (DAC) component of a standard analog modem. Modulation in modems allows digital binary information to be received from a serial interface on a computer and modulated into sound waves for transmission over the voice-grade PSTN.

Modulation. Examples of different types of modulation.

Common types of signal modulation that are used for converting digital to analog signals include

Frequencies Used in FSK Modulation

FSK Frequency

Meaning

1070 Hz

Transmitted binary 1

1270 Hz

Transmitted binary 0

2025 Hz

Received binary 0

2225 Hz

Received binary 1
QAM Encoding of Binary Value Pairs

QAM Phase Value

Binary Information Represented

45 phase shift

11

135 phase shift

10

225 phase shift

01

315 phase shift

00

Notes

For modulation (line coding) over digital transmission lines, various techniques can be employed, depending on the networking technology under consideration. These include

See Also Asymmetric Digital Subscriber Line (ADSL) , Digital Subscriber Line (DSL) ,electromagnetic interference (EMI) ,line coding , Public Switched Telephone Network (PSTN), signal, Very-high-rate Digital Subscriber Line (VDSL)

MOF

Stands for Microsoft Operations Framework, a service from Microsoft Corporation that provides technical guidance on Microsoft technologies to large enterprises.

See Also Microsoft Operations Framework (MOF)

MOSPF

Stands for Multicast Open Shortest Path First, a dense mode multicast routing protocol.

See Also Multicast Open Shortest Path First (MOSPF)

MOUS

Stands for Microsoft Office User Specialist, a Microsoft certification for users of Microsoft Office.

See Also Microsoft Office User Specialist (MOUS)

moving files

Changing the directory in which a file is stored.

Overview

On Microsoft Windows platforms, files can be moved using a graphical user interface (GUI) tool such as Windows Explorer or from the command prompt using the Move command. Some inheritance issues are associated with moving files on Windows NT and Windows 2000 platforms that use the NTFS file system (NTFS). In particular, the effect of moving a file depends on whether the file is moved within a volume or to a different volume:

Notes

If you move an object within Active Directory directory service on a Windows 2000-based network, the object loses any permissions that were assigned to the organizational unit (OU) in which it resided and inherits the permissions assigned to the OU to which it is moved. Also, any permissions that were explicitly assigned to the object remain in effect after the object is moved. For example, you might move an object in Active Directory if a user has changed departments and you have to move the user object from the Sales OU to the Marketing OU. Use the Active Directory Users and Computers snap-in for Microsoft Management Console (MMC) to move objects in Active Directory.

See Also copying files

Mozilla

An open-source Web browser.

Overview

Mozilla is an open-source project aimed at developing a free, standards-based Web browser. Mozilla is based on Netscape Navigator, a popular Web browser platform developed by Netscape in the mid-1990s which was itself based on the earlier NCSA Mosaic browser. Mozilla will run on a variety of operating system platforms, including Microsoft Windows, Apple Macintosh, and Linux. The development of the Mozilla browser is being coordinated by Mozilla.org, an open source organization.

For More Information

Visit mozilla.org for details on Mozilla.

See Also Internet Explorer ,open source ,Web browser

MP

Stands for Multilink Point-to-Point Protocol, a wide area network (WAN) protocol for aggregating multiple Point-to-Point Protocol (PPP) connections.

See Also Multilink Point-to-Point Protocol (MPPP)

MPLS

Stands for Multiprotocol Label Switching, a protocol for efficiently routing traffic across a large Internet Protocol (IP) network such as the Internet.

See Also Multiprotocol Label Switching (MPLS)

MPOA

Stands for Multiprotocol over ATM, a method for forwarding Internet Protocol (IP) packets over Asynchronous Transfer Mode (ATM) networks.

See Also Multiprotocol over ATM (MPOA)

MPPP

Stands for Multilink Point-to-Point Protocol, a wide area network (WAN) protocol for aggregating multiple Point-to-Point Protocol (PPP) connections.

See Also Multilink Point-to-Point Protocol (MPPP)

MSAU

Another acronym that stands for Multistation Access Unit, a wiring concentrator (passive hub) used in Token Ring networks. An alternative to MAU, the most common acronym for this term.

See Also Multistation Access Unit (MAU or MSAU)

MS-CHAP

Stands for Microsoft Challenge Handshake Authentication Protocol, an encrypted authentication scheme for Point-to-Point Protocol (PPP) sessions.

See Also Microsoft Challenge Handshake Authentication Protocol (MS-CHAP)

MSDN

Stands for Microsoft Developer Network, Microsoft Corporation's portal and support program for their developer community.

See Also Microsoft Developer Network (MSDN)

MS-DOS

Stands for Microsoft Disk Operating System, an operating system created by Microsoft Corporation in 1981 for the first IBM personal computer (PC).

See Also Microsoft Disk Operating System (MS-DOS)

MS-DOS mode

An operating system mode for some versions of Microsoft Windows that is compatible with running MS-DOS applications.

Overview

MS-DOS mode is a special operating system mode in Windows 95, Windows 98, and Windows Millennium Edition (Me) that allows applications to access hardware directly. This functionality is sometimes needed in order to run older MS-DOS applications (especially games) on these versions of Windows. You can also use MS-DOS mode when you have to run stubborn MS-DOS programs that will not run within an MS- DOS window on these versions of Windows 95.

To switch to MS-DOS mode from Windows, choose Shutdown from the Start menu, and then choose Restart In MS-DOS Mode. The Windows graphical user interface (GUI) vanishes, leaving only an MS-DOS prompt, which allows you to run a single MS-DOS-based application. A stub of the Windows operating system remains in memory and restores the Windows GUI when you type exit at the MS-DOS prompt.

See Also Microsoft Disk Operating System (MS-DOS) ,Microsoft Windows

M-Services

Stands for Mobile Services Initiative, a proposed set of standards for broadband cellular Internet access.

See Also Mobile Services Initiative (M-Services)

MSP

Stands for Management Service Provider, a company that manages the information technology (IT) infrastructure for other businesses.

See Also Management Service Provider (MSP)

MSF

Stands for Microsoft Solutions Framework, a training program developed by Microsoft Corporation for large enterprises that provides guidance on planning, building, and deploying solutions based on Microsoft products and technologies.

See Also Microsoft Solutions Framework (MSF)

MSR

Stands for Microsoft Research, a branch of Microsoft Corporation concerned with advanced research in computer science and technologies.

See Also Microsoft Research (MSR)

MTC

Stands for Microsoft Technology Center, a state-of-the-art training center for learning about Microsoft technologies and solutions.

See Also Microsoft Technology Center (MTC)

MTQP

Stands for Message Tracking Query Protocol, an emerging protocol for tracking e-mail messages over the Internet.

See Also Message Tracking Query Protocol (MTQP)

MTSO

Stands for Mobile Transport Serving Office, a component of a cellular communications system.

See Also Mobile Transport Serving Office (MTSO)

MTU

Stands for multitenant unit, a building with many tenants, such as a skyscraper.

See Also multitenant unit (MTU)

Multicast Address Dynamic Client Allocation Protocol (MADCAP)

An extension to the Dynamic Host Configuration Protocol (DHCP) for dynamic assignment of multicast addresses.

Overview

MADCAP is an extension to DHCP that offers dynamic assignment and configuration of Internet Protocol (IP) multicast addresses on Transmission Control Protocol/Internet Protocol (TCP/IP) networks. The MADCAP extensions to DHCP are described in RFC 2131, which makes the earlier RFC 1541 obsolete. MADCAP is supported by the Microsoft Windows 2000 operating system platform.

Implementation

In the Windows 2000 and Windows .NET Server implementations, MADCAP allows DHCP servers to include class D multicast IP addresses in separate DHCP scopes. These class D addresses can range from 224.0.0.0 through 239.255.255.255, although some class D addresses are permanently reserved for specific multicast groups and others can be assigned as desired. MADCAP dynamically assigns available class D addresses to clients who need to temporarily join a multicast group in order to receive a multicast transmission. The DHCP clients must also support MADCAP in order to request and receive multicast addresses from the MADCAP service (client machines running Windows 2000 and Windows .NET Server also support MADCAP).

MADCAP operates separately from DHCP, and a client that has an IP address from a DHCP server (or has a static IP address that has been manually assigned to it) can request from the MADCAP service an additional multicast address so that it can receive a multicast transmission over the network. A client can lease only one unicast address, but it can request and obtain multiple multicast addresses and belong to multiple multicast groups simultaneously.

Notes

If a client does not support MADCAP, it cannot request addresses from the MADCAP service, but it might be able run multicast-enabled applications such as Windows Media Player to receive multicast transmissions over a network.

MADCAP was formerly known as the Multicast Dynamic Host Configuration Protocol (MDHCP).

See Also Dynamic Host Configuration Protocol (DHCP) , unicasting

Multicast Dynamic Host Configuration Protocol (MDHCP)

The former name for what is now called Multicast Address Dynamic Client Allocation Protocol (MADCAP).

See Also Multicast Address Dynamic Client Allocation Protocol (MADCAP)

multicasting

Also called IP multicasting, a method for sending one Internet Protocol (IP) packet simultaneously to multiple hosts.

Overview

Most communications between two hosts on an IP network are performed using one of two methods:

Unfortunately, neither of these two IP transport methods is suited to a rapidly growing segment of Internet traffic-multimedia presentations and videoconferencing. For example, consider one host transmitting an audio or video stream and 10 hosts that want to receive this stream. If unicasting were used, the sending host would need to transmit 10 separate IP packets, one for each receiving host. As the number of receiving hosts scales upward, this scenario rapidly results in saturation of network bandwidth. On the other hand, if broadcasting were used by the sending host, then only one IP packet (a broadcast packet) would need to be transmitted, not just the 10 hosts but all hosts on the network would receive and process the packet. The result is that network hosts would incur unnecessary processing, which is clearly also an undesirable situation.

The solution to this is multicasting-the sending host transmits a single IP packet with a specific multicast address, and the 10 hosts simply need to be configured to listen for packets targeted to that address to receive the transmission. Multicasting is thus a point-to-multipoint IP communication mechanism that operates in a connectionless mode-hosts receive multicast transmissions by "tuning in" to them, a process similar to tuning in to a radio station. Multicasting is supported by Ethernet, Token Ring, Fiber Distributed Data Interface (FDDI), and Asynchronous Transfer Mode (ATM) networking architectures.

Architecture

IP multicast addresses (also known as group addresses) are IP addresses belonging to class D, and therefore fall within the range 224.0.0.0 through 239.255.255.255. IP multicast addresses thus have their four high-order bits set to 1110 when viewed in binary notation. In Classless Inter-Domain Routing (CIDR) notation, the block of IP multicast addresses is expressed as 224.0.0.0/4.

Multicast addresses come in two types:

Examples of Reserved IP Multicast Addresses

Address

Description

224.0.0.0

Base address (not used)

224.0.0.1

All hosts on the local subnet

224.0.0.2

All routers on the local subnet

224.0.0.5

All Open Shortest Path First (OSPF) version 2 routers on the network

224.0.0.6

Designated Open Shortest Path First (OSPF) version 2 routers on the network

224.0.0.9

Routing Information Protocol (RIP) version 2 group address

224.0.0.18

Internet Group Management Protocol (IGMP) group address

224.0.1.1

Network Time Protocol (NTP)

224.0.1.24

WINS server group address

224.0.1.75

Session Initiation Protocol (SIP) group address

224.0.12.0 to 224.0.12.63

Used by MSNBC

224.0.18.0 to 224.0.18.255

Used by Dow Jones

IP hosts can support multicasts at one of three levels:

Microsoft Windows 2000 supports Level 2 multicast functionality for all forms of IP traffic.

Implementation

Membership in a multicast group is dynamic-hosts can join or leave a group as they choose. To join a multicast group, a host sends a message to the all hosts multicast group 224.0.0.1. This informs routers on the local subnet that the host wants to join a specific multicast group. Routers then periodically send queries to the same multicast group to ensure hosts are still interested in belonging to multicast groups. Hosts can belong to more than one multicast group simultaneously, and multicast groups can span multiple subnets.

When the host wants to leave a group, it sends a message to the All Routers multicast group 224.0.0.2. These messages between hosts and routers are sent using the Internet Group Management Protocol (IGMP).

Routers must be multicast-enabled to support multicasting. Multicast routers communicate with each other using multicast routing protocols such as Multicast Open Shortest Path First (MOSPF) and Protocol Independent Multicast (PIM).

For More Information

Visit the IP Multicast Initiative at www.ipmulticast.com

See Also broadcasting , Class D ,classless interdomain routing (CIDR) ,Internet Group Management Protocol (IGMP) ,Internet Protocol (IP) ,IP address , router, unicasting

Multicast Open Shortest Path First (MOSPF)

A dense mode multicast routing protocol.

Overview

Multicast Open Shortest Path First (MOSPF) is a protocol that allows multicast-enabled routers to communicate with each other in order to build multicast forwarding tables. These tables contain trees of information specifying which subnets have hosts that belong to different multicast groups. This allows hosts belonging to a specific group to receive multicast transmissions directed toward that group.

MOSPF is designed to be used within a single routing domain or autonomous system (AS). Although it is classified as a dense mode multicast routing protocol, MOSPF does not work well for networks that have large numbers of active multicast groups due to the large amount of overhead involved in recalculating distribution trees when multicast groups are added or removed from the network. Performance is best for MOSPF when only a few multicast groups are active, and, as a result, MOSPF is not widely used for multicasting over the Internet.

MOSPF is based on the unicast Open Shortest Path First (OSPF) routing protocol and is defined in RFC 1584. Cisco routers do not support MOSPF.

See Also autonomous system (AS) , Open Shortest Path First (OSPF)

multicast routing

Routing of multicast packets across an internetwork from the sending host to the receiving group of hosts.

Overview

Ordinary routing is unicast routing-that is, the routing of unicast packets. Unicast routing involves forwarding Internet Protocol (IP) packets from the sending host to a specific, globally unique receiving host. Routing involves forwarding the packet across multiple subnets to its final destination, the number of subnets crossed being the number of hops required for the packet to reach its destination. From the point of view of unicasting, subnets are well-defined entities connected by routes.

Multicast routing is different: multicast packets are forwarded to multicast-enabled routers and these routers are responsible for ensuring that hosts belonging to the multicast group specified by the destination address receive the transmitted packet. Group members may be located in a single subnet or in many different subnets and may be concentrated either in a small portion of the larger internetwork or scattered all over the place. To ensure that multicast packets are not forwarded to subnets where no interested hosts reside and to ensure that they are forwarded from routers on noninterested subnets to other routers further downstream, multicast routing protocols are used to allow multicast-enabled routers to communicate with each other and exchange information.

Information about which subnets interested hosts reside on is maintained in multicast forwarding tables stored on multicast-enabled routers. An entry in this table indicates that there is at least one interested host (a host belonging to the multicast group to which the transmission is directed) in the subnet mapped to that entry. Multicast routing protocols generally store their multicast forwarding tables as hierarchical structures called multicast trees. A tree defines the distribution of multicast packets from the sending host to all receiving hosts within its multicast group. Mulitcast trees are created and maintained by processes known as grafting and pruning and can be either source trees which find the shortest path from the sending host to each potential receiving host, or shared trees which merge source trees into a single tree for each specific group. Multicast traffic is prevented from being sent to subnets where no interested hosts reside by a process known as scooping.

See Also multicasting ,multicast routing protocol routing, unicasting

multicast routing protocol

Any protocol used for communication between multicast-enabled routers.

Overview

Multicasting on an Internet Protocol (IP) internetwork relies on special multicast-enabled routers. These routers enable hosts to register themselves for sending or receiving multicast transmissions. Multicast routers communicate with each other using multicast routing protocols in order to create multicast forwarding tables (multicast trees) that ensure that each host that joins a multicast group receives the transmission directed toward that group.

Types

There are two basic kinds of multicast routing protocols:

See Also Core-Based Trees (CBT) , Distance Vector Multicast Routing Protocol (DVMRP) , Protocol Independent Multicast-Dense Mode (PIM-DM), Protocol Independent Multicast-Sparse Mode (PIM-SM), routing protocol

multihoming

The process of having more than one network interface on a device.

Overview

Networking devices that have more than one interface are called multihomed devices. A router is a good example of a multihomed device because it has several network interfaces, each usually connected to a different subnet.

Computers can also be multihomed by giving them multiple interfaces to a network. This is true even if the computer has only a single physical interface (network interface card, or NIC). For example, Microsoft Windows 2000 allows you to assign more than one Internet Protocol (IP) address to a single physical interface, resulting in multiple virtual interfaces and making the machine a multihomed host. When a computer is connected to both a local area network (LAN) using a NIC and the Internet using a dial-up connection, this can also be considered multihoming since the dial-up connection is also an interface (a wide area network [WAN] interface typically running Point-to-Point Protocol [PPP]), though the term multihoming is not always used in this scenario.

The most common way of multihoming a computer, however, is to install two or more physical interfaces (NICs) and assign separate IP addresses to each NIC. Windows 2000 supports all of these methods of multihoming. Note that the first method of using multiple virtual interfaces is sometimes called multinetting and is considered by some not to be true multihoming.

Uses

Multihoming has a number of possible uses:

Issues

Some issues can arise with multihomed hosts:

Notes

The term multihoming is also used in a different sense in the modern enterprise: using several Internet Service Providers (ISPs) to connect your network to the Internet. A typical multihoming scenario would see two leased lines joining your network to the Internet. Each line is leased from a different ISP, and one line is used to bring IP traffic into your network while the other is used to carry traffic out of the network. The motive for this kind of multihoming is to ensure redundant connection for enterprise data centers to the high-speed backbone of the Internet through different points of presence (POPs) belonging to different carriers. This is desirable for companies that view the Internet as a mission- critical resource and cannot afford any downtime.

To implement Internet multihoming like this generally requires high-end routers and an understanding of complex protocols, such as Border Gateway Protocol (BGP), that are used on the Internet's backbone. Often, the challenge in this scenario is to get the different ISPs to cooperate with one another to ensure your network does not have IP address space problems. This is because a multihomed network needs to have its own autonomous system number (ASN) assigned to it by an Internet registry such as the American Registry for Internet Numbers (ARIN). Many large ISPs usually will not agree to this kind of multihoming arrangement unless a minimum of a T3 connection is employed, which means that only large companies such as IBM, Hewlett-Packard Company, Amazon.com, and eBay can afford to implement this kind of multihoming. Nevertheless, the steady rise in the number of ASNs being allocated by ARIN over the last few years indicates the rise in popularity of Internet multihoming among large enterprises and e-business companies. A cheaper alternative to multihoming for enterprises that need redundancy is simply to lease several T1 or T3 lines from the same carrier and inverse multiplex them into a single connection.

See Also American Registry for Internet Numbers (ARIN) ,autonomous system number (ASN) ,Border Gateway Protocol (BGP) ,Dynamic Host Configuration Protocol (DHCP) ,interface ,inverse multiplexing ,network interface card (NIC) ,router

Multilayer Switching (MLS)

A local area network (LAN) switching term that can have several meanings.

Overview

The term Multilayer Switching (MLS) can have several meanings within the context of LAN switching. Specifically, it can refer to

See Also Ethernet switch ,Layer 2 switch ,Layer 3 switch ,Layer 4 switch ,Open Systems Interconnection (OSI) reference model ,router

Multilink Frame Relay (MFR)

A new frame relay aggregation technology.

Overview

Multilink Frame Relay (MFR) is a new standard from the Frame Relay Forum that allows multiple frame relay links to be aggregated using inverse multiplexing. The goal of MFR is to make frame relay an attractive technology in the market niche between T1 and T3 lines. Enterprises using multiplexed T1 lines that require more bandwidth often balk at the cost of upgrading to T3, which offers speeds of 45 megabits per second (Mbps) compared to T1's 1.5 Mbps. MFR is designed to provide a lower-cost option that can provide bandwidth intermediate between T1 and T3.

Implementation

MFR is implemented at the service provider end using MFR-enabled carrier-class frame relay switches and at the customer premises end using MFR-enabled Integrated Access Devices (IADs). Multilink Point-to-Point Protocol (MPPP) service providers include competitive local exchange carriers (CLECs), building local exchange carriers (BLECs), and large Internet service providers (ISPs). Service providers generally target midsize businesses and multitenant units (MTUs) for deploying MFR.

MFR is often used in conjunction with MPPP, a wide area network (WAN) protocol that allows scalable bandwidth allocation using dynamic bonding of multiple Point-to-Point Protocol (PPP) links. MPPP devices can aggregate up to 128 separate frame relay links into a single logical pipe, with individual links ranging from DS-0 to T1.

See Also Building-centric Local Exchange Carrier (BLEC) , Competitive Local Exchange Carrier (CLEC) ,DS-0 ,frame relay ,Integrated Access Device (IAD) ,Internet service provider (ISP) , Point-to-Point Protocol (PPP), T-carrier

Multilink Point-to-Point Protocol (MPPP)

A wide area network (WAN) protocol for aggregating multiple Point-to-Point Protocol (PPP) connections.

Overview

Multilink Point-to-Point Protocol (MPPP) can be used to bundle together multiple physical PPP links into a single logical link. This is accomplished by using inverse multiplexing on each end of the link, at both the customer premises, and at the service provider (for an MPPP connection to be established, both ends of the connection must support MPPP). MPPP's purpose is usually to provide greater bandwidth for a WAN connection than a single PPP link can provide, but since MPPP is a dynamic process, it can also be used to better utilize bandwidth and even load-balance between connections. The advantages of using MPPP on WAN links include increased bandwidth, dynamic bandwidth allocation, reduced latency, and fault tolerance. MPPP's main disadvantage is that it is difficult to support callback for remote access servers.

MPPP is an extension to the industry-standard PPP and is defined in RFC 1990. MPPP is supported by most access servers and routers and also by the Microsoft Windows 2000 operating system. Multilink Point-to- Point Protocol is variously abbreviated as MPPP, MP, or MLP, and is abbreviated occasionally (and wrongly) as MPP or even MLPPP. Sometimes it is simply called Multilink Protocol for short!

Uses

A popular use of MPPP is in Integrated Services Digital Network (ISDN) networking where it can be used to either bond (combine) the two 64-kilobits per second (Kbps) B channels of an ISDN-BRI interface into a single 128-Kbps logical channel or aggregate two or more BRI interfaces together for even greater throughput.

Multilink Point-to-Point Protocol (MPPP). Using MPPP to bond two PPP links together.

MPPP was the first industry-standard nonproprietary method for ISDN bonding, and as a result, it is popular where ISDN technologies are still used. Because MPPP is a dynamic protocol, it can also be used to dynamically bring online the second B channel whenever the bandwidth of a single channel is insufficient and then dynamically release the second channel when it is no longer required. This is done by configuring a traffic threshold above which the second channel will automatically be activated. MPPP is supported by most ISDN terminal adapters.

A newer use of MPPP is in ISDN Digital Subscriber Line (IDSL), a form of Digital Subscriber Line (DSL) technology based on ISDN signal coding. By installing an MPPP-enabled IDSL router at the customer premises and a similar router at the service provider's point of presence (POP), up to four 144 kilobits per second (Kbps) IDSL links can be multiplexed into a single 576 Kbps link. MPPP IDSL is a technology pioneered by Netopia and is intended mainly for corporate WAN use and not for residential Internet access. Using MPPP IDSL eliminates the need for installing a costly Digital Subscriber Line Access Multiplexer (DSLAM) at the customer premises and is ideal for environments such as remote telco terminals that are too limited in size to house DSLAMs.

Implementation

MPPP can be implemented over a variety of interfaces, including asynchronous dial-up modem connections, Integrated Services Digital Network (ISDN), and even synchronous connections. MPPP is negotiated during the Link Control Protocol (LCP) portion of a PPP authentication session.

MPPP defines procedures for splitting a PPP data stream into packets, sequencing the packets into time slots, transmitting them over a logical data link, and reassembling them at the receiving station. MPPP works by inverse multiplexing data frames from multiple client PPP connections into a single PPP link, and then demultiplexing the link to recreate the individual connections at the service provider's router.

MPPP supports two kinds of PPP authentication: Challenge Handshake Authentication Protocol (CHAP) and Password Authentication Protocol (PAP).

Notes

A proprietary extension to MPPP supported by some vendors is called Multichassis Multilink Point-to-Point Protocol (MMP), which allows MPPP connections to be aggregated across multiple routers and network access servers (NASs) in a way that is transparent to the dial-up MPPP client. In other words, the client initiates an MPPP session but is actually connected to several MMP-enabled NASs at the ISP instead of only one NAS, as in the usual scenario. MMP enables the data stream to be split, sequenced, and recombined at several different points to provide a single logical connection between the client and the ISP.

Another proprietary extension to MPPP is called Multichannel PPP (MPP), which in addition to inverse multiplexing of PPP links also supports session and bandwidth management functions, including the dynamic addition or removal of channels without the need to reinitialize the link. Both the client and the server must support MPP for this to work.

See Also Challenge Handshake Authentication Protocol (CHAP) ,Digital Subscriber Line (DSL) ,Digital Subscriber Line Access Multiplexer (DSLAM) ,Integrated Services Digital Network (ISDN) ,inverse multiplexing ,ISDN Digital Subscriber Line (IDSL) ,ISDN fallback adapter ,Link Control Protocol (LCP) ,Password Authentication Protocol (PAP) ,Point-to-Point Protocol (PPP) ,wide area network (WAN)

multimaster replication

Replication between peers.

Overview

Replication is an important aspect of many kinds of network services. For example, in network directory services, replication ensures that each directory server has an up-to-date version of directory information for the network. Replication between directory servers can take place in two ways: master/slave and multimaster replication.

In master/slave replication, the master server maintains the master copy of directory information and replicates this information with slave servers through periodic updates. Directory information can usually only be modified directly on the master server-slave servers usually have read-only versions of directory information. Master/slave replication is reliable but scales poorly and has a single point of failure-should the master server go down, the directory cannot be updated unless a slave server is promoted to the role of master. Master/slave replication is used by the Windows NT Directory Services (NTDS) of the Microsoft Windows NT 4.0 operating system, where master servers are called primary domain controllers (PDCs) and slaves are backup domain controllers (BDCs). In Windows NT, directory information is stored on domain controllers in the Security Account Manager (SAM) database, and although the PDC for each domain has a writable copy of the SAM database, the BDCs for that domain have read-only copies of the database.

In multimaster replication, however, there is no master directory server-all directory servers are peers of one another. Multimaster replication scales well and has no single point of failure, but it must be implemented properly using time stamping of updates and time synchronization between servers to ensure that replication occurs properly if collisions are to be avoided. A collision occurs when an object in Active Directory directory service has the same attribute modified almost simultaneously on two different domain controllers, and collisions are resolved using timestamps as a tie- breaking mechanism. Active Directory in Microsoft Windows 2000 uses multimaster replication to replicate directory information between all domain controllers in a domain. Multimaster replication, in fact, means that all domain controllers are "master" domain controllers-in other words, applications and users having sufficient privileges can update directory information on these servers directly.

The advantages of the multimaster replication method supported by Windows 2000 over the PDC/BDC master/slave replication method of Windows NT include the following:

See Also Active Directory ,directory

multimode fiber-optic cabling

A type of fiber-optic cabling that can carry multiple signals simultaneously.

Overview

Multimode fiber-optic cabling allows large amounts of information to be transmitted over a single strand of fiber. This is accomplished by transmitting multiple light signals simultaneously through specially constructed fiber. This enables multimode fiber to sustain a much greater bandwidth than single-mode fiber, a type of fiber that supports only one light signal at a time.

The disadvantage of multimode cabling is that multimode transmission is more complex and incurs greater risk of signal loss. As a result, multimode fiber runs cannot be as long as single-mode fiber, which can carry signals for several kilometers without degradation. So although multimode fiber can carry many times more bandwidth than single-mode fiber, single-mode fiber can generally carry signals up to 50 times farther than multimode.

Implementation

Multimode fiber is fiber-optic cabling that has a glass core whose index of refraction varies in a specific fashion with the distance from the core axis. This variation in index of refraction is done to ensure that multiple separate light signals can be effectively transported down the fiber without interference or signal loss. The variation is implemented in one of two ways:

Multimode fiber is available with different core diameters, typically 50, 62.5, and 100 microns. Line drivers transmitting signals through multimode cables generally use light-emitting diodes (LEDs) to generate their signals.

Uses

Because of the length restrictions for multimode fiber, this type of cabling is generally restricted in use to small cable runs such as switch-switch and switch-server links and backbone connections within buildings (intrabuilding connections). Single-mode fiber is necessary for long cable runs such as inter-building connections and distributed campus backbones.

Multimode cable lengths are restricted to various distances depending on the networking technology used. For example, in Gigabit Ethernet (GbE), multimode fiber of the 50 and 62.5 micron size are supported (62.5 is preferred) with maximum cable run lengths supported being 1800 feet (550 meters). If this limit is exceeded, the light traveling along different paths through the fiber can produce a condition called modal dispersion, which results in parts of the signal arriving at unexpected times at the end station. This condition can quickly degrade the signal's quality, causing it to be unrecognizable so that communication is impossible. Note that older forms of Ethernet such as 10BaseT and Fast Ethernet support multimode links up to 6600 feet (2000 meters) in length due to their slower signaling speed.

See Also cabling ,Ethernet ,Fast Ethernet ,fiber-optic cabling ,Gigabit Ethernet (GbE) ,single-mode fiber-optic cabling

multiple master domain model

A strategy for deploying large networks based on Microsoft Windows NT.

Overview

The multiple master domain model represents a Windows NT domain model in which all user and group accounts for the enterprise reside in two or more account domains that trust one another using two-way trusts. Network resources for the enterprise reside in separate resource domains, and each resource domain trusts every account domain in the enterprise. The goal of this structure is to ensure that users anywhere in the enterprise can access network resources anywhere in the enterprise provided they have suitable permissions.

Multiple master domain model. Using the multiple master domain model to deploy Windows NT for a large enterprise.

Large enterprises may use the multiple master domain model either because of limitations on the recommended number of users in a Windows NT domain (about 20,000) or because their company is spread across several geographically separate locations.

Advantages and Disadvantages

The advantages of the multiple master domain model are that it is scalable to any number of user accounts and that resource domains in this model manage their own resources. Among the disadvantages are the fact that the master domain model is complex to set up and administer and that multiple local groups must be created in each resource domain.

Notes

Enterprises can simplify network administration considerably by upgrading their large Windows NT-based networks to Windows 2000. Windows 2000 overcomes the limitations of Windows NT's multiple master domain model by supporting the storage of up to 10 million objects in Active Directory directory service and by allowing a single domain to be partitioned into multiple sites with intersite directory replication traffic being scheduled for times of low wide area network (WAN) usage. As a result, it is often the best approach and a relatively straightforward process to consolidate a large Windows NT-based network having multiple master domains and resource domains into a single large Windows 2000 domain having multiple sites. In this approach you can mirror the administrative structure of your former network by creating a hierarchy of organizational units (OUs) within Active Directory and assigning suitable permissions for these OUs to users and groups.

If you want to maintain the existing administrative structure of your company more closely, however, you can take a different approach in your migration. You can migrate each Windows NT master or resource domain to a distinct Windows 2000 domain. Here each master domain would become the root of a domain tree, and the resource domains that trusted them become child domains joined by two-way transitive trusts to their parent domain. This results in one new domain tree for each former master domain. You then establish two-way transitive trusts between the root domains of each tree, forming a domain forest. All users in your enterprise thus gain access to resources anywhere on the network.

See Also Active Directory , complete trust model , organizational unit (OU), single domain model

Multiple Virtual Storage (MVS)

A legacy IBM mainframe operating system.

Overview

Multiple Virtual Storage (MVS) was originally developed for the System/360 series of IBM mainframes. The name comes from the fact that MVS uses virtual memory (storage) for sharing memory among multiple applications.

MVS is still used on many legacy mainframes found in government, finance, and industry for running accounting and payroll systems, large databases, and other legacy applications. MVS is mainly used to run applications written in COBOL, but it supports other languages, including FORTRAN, PL/1, and CICS.

MVS has been supplanted by the OS/390 operating system, which runs on IBM's System/390 mainframe models.

See Also mainframe

multiplexer (MUX)

A device that combines multiple data streams into a single stream.

Overview

Multiplexers (MUXes) are used in wide area networks (WANs) to make optimum use of available bandwidth in leased lines. A multiplexer allows data streams from several pieces of data terminal equipment (DTE) to be multiplexed (combined) into a single stream that can be transmitted over the leased line. In addition, multiplexers allow leased lines to carry voice, video, and data traffic simultaneously.

An inverse multiplexer, or IMUX, takes a single input data channel from a DTE, such as a router, and spreads it to several digital transmission lines. A demultiplexer is simply a MUX at the remote end of a multiplexed communication system. A modular MUX allows you to expand your wide area network (WAN) link as your network grows by adding modules to provide additional V.35, RS-232, or RS-530 input ports.

Implementation

Multiplexers can save companies costs by eliminating the need for extra line termination devices such as Channel Service Unit/Data Service Units (CSU/DSUs). For example, a T1 MUX typically has several V.35 interfaces that can accept incoming data from switches or routers connected to your local area networks (LANs), plus a DSX-1 interface for connecting to a channel bank or private branch exchange (PBX) system. The MUX takes this serial data input and voice input and multiplexes them together into a single data stream using an integrated CSU/DSU and then transmits this stream over the T1 line, letting you use your T1 line for voice and data combined. You can use a similar device to multiplex data and voice for transmission over a public frame relay carrier network.

Multiplexer (MUX). Examples of using a T1 multiplexer and a fiber-optic MUX.

These T1 MUXes often offer advanced management functions. For example, a T1 MUX can usually be remotely managed using the Simple Network Management Protocol (SNMP), and it can be configured to send a trap to an SNMP management console whenever a problem occurs with the MUX. Some MUXes are also used in mainframe environments for connecting remote terminals to async hosts without the need for individual cabling to each terminal. These kinds of MUXes are generally used in pairs and utilize time-division multiplexing (TDM). Examples include the following:

Another type of multiplexer is the Ethernet MUX, which can multiplex signals from several ports from an Ethernet switch and transmit the multiplexed signal over a single fiber-optic cable at distances up to 1.25 miles (2 kilometers). Ethernet MUXes are also typically used in pairs.

See Also Channel Service Unit/Data Service Unit (CSU/DSU) , data terminal equipment (DTE) ,inverse multiplexing , Simple Network Management Protocol (SNMP), T1, T-carrier

multiplexing

Combining multiple data streams into a single stream.

Overview

Multiplexing is used in networking and telecommunications to aggregate data streams from different sources into a single channel, typically for transmission over a wide area network (WAN) connection. For example, signals from several data terminal equipment (DTE) can be combined into a single signal using a multiplexer and transmitted over a leased line such as a T1 line. For information to be transmitted successfully, the leased line must have a bandwidth equal to or greater than the combined bandwidth of the signals from the various DTE.

Implementation

Several methods of multiplexing signals are commonly used in telecommunications and networking. Each method is suited to a specific form of communication. Some more common methods used include

See Also Asynchronous Transfer Mode (ATM) , dense wavelength division multiplexing (DWDM) ,frame relay ,frequency-division multiplexing (FDM) ,inverse multiplexing ,leased line , statistical multiplexing (STM), time-division multiplexing (TDM)

multipoint

Short for point-to-multipoint, communication from a single sending station to multiple receiving stations.

See Also point-to-multipoint

Multipoint Multichannel Distribution Service (MMDS)

A fixed wireless broadband networking architecture.

Overview

Also known as fixed wireless, Multipoint Multichannel Distribution Service (MMDS) is a new broadband wireless service designed to compete with Digital Subscriber Line (DSL) and cable modem technologies. It can be used to provide enterprise with high-speed wireless wide area network (WAN) connections and for high-speed Internet access.

MMDS is a serious contender in the enterprise WAN arena and is a low-cost alternative to expensive leased lines such as T1 and fractional T1 lines. MMDS is generally considered superior to DSL and cable modem technologies for enterprise use for several reasons:

Implementation

MMDS operates in the 2.1, 2.5, and 2.7 gigahertz (GHz) portions of the electromagnetic spectrum. Typical speeds for MMDS are 384 kilobits per second (Kbps) to 1 megabit per second (Mbps) for downstream transmission and 384 Kbps to 512 Kbps upstream.

In a typical MMDS scenario, a base station with antenna is set up at a high location (building rooftop or hill) in or near an urban area. The base station provides coverage for customer sites within a 22-mile (35-kilometer) radius. Customer sites must have a clear line of sight with the base station's antenna in order for MMDS to work properly.

Marketplace

Currently, the two major players in the MMDS arena are Sprint Corporation and MCI/Worldcom, two inter- exchange carriers (IXCs). MMDS is being rolled out by these companies in major urban areas around the United States. A major producer of MMDS equipment is Vyyo Wireless Systems.

Prospects

The main issue with MMDS is the clear line of sight (LoS) requirement for transmission. One way of working around this is to implement orthogonal frequency division multiplexing (OFDM) on MMDS, which helps overcome the LoS, reach, and antenna size issues relating to MMDS.

See Also broadband Internet access ,cable modem ,Digital Subscriber Line (DSL) ,frame relay ,Orthogonal Frequency Division Multiplexing (OFDM) ,T-carrier ,wide area network (WAN) ,wireless networking

multiport repeater

An older name for a hub, a device used to connect shared Ethernet segments into a single LAN.

See Also hub

Multiprotocol Label Switching (MPLS)

A protocol for efficiently routing traffic across a large Internet Protocol (IP) network such as the Internet.

Overview

Multiprotocol Label Switching (MPLS) is an emerging standard from the Internet Engineering Task Force (IETF) designed for implementation on high-speed backbone routers in large IP networks such as the Internet. MPLS is an outgrowth of several proprietary vendor-based solutions to the problem of routing traffic through high-speed switched backbones. The most prominent of these proprietary solutions is the tag switching technology developed by Cisco Systems, from which much of MPLS was derived. Other companies whose technologies have contributed to the development of MPLS include Ipsilon Networks (now part of Nokia), Cascade Communications, 3Com Corporation, IBM, and Cabletron Systems.

MPLS is designed to bring some of the advantages of circuit-switched networks to switched IP networks. These advantages include predictable delay and latency, the ability to reserve bandwidth, and different levels of Quality of Service (QoS).

Implementation

The name MPLS defines two aspects of this protocol:

An MPLS label is simply a four-octet binary number. The first 20 bits of the label uniquely identify the Forwarding Equivalence Class (FEC) to which the packet belongs, while the remaining bits specify the time to live of the packet and other information. Each data flow through the internetwork is mapped to an FEC-in other words, all packets belonging to the same FEC are handled similarly by MPLS-enabled routers. Multiple labels can be attached to each packet, which enables MPLS to support features such as tunneling and even stacked tunnels. A sequence of labels or a label stack is called a Label-Switched Path (LSP).

MPLS-enabled routers are called Label Switching Routers (LSRs), and for MPLS to work properly it must be supported by routers all across the internetwork. LSRs forward packets by examining their LSP and switching them accordingly, a process that can be much faster than traditional IP routing. Packets are introduced into an MPLS domain using a Label Edge Router (LER) connected to both the local network and the backbone network. Once the packet is on the backbone, it is label-switched to the LER adjacent to the destination network, at which point it is routed using traditional IP routing to its final destination. LSRs communicate with each other using the Label Distribution Protocol (LDP) to let each other know what labels they are assigning to different traffic flows.

Uses

Because MPLS is a complex protocol, it is unlikely to be used extensively in the enterprise. Instead, service providers and telcos are likely to deploy MPLS on their ATM backbones to increase the efficiency with which they carry IP traffic. MPLS offers an alternative to permanent virtual circuits (PVCs) commonly used in telco ATM networks. Using MPLS, ATM switches become routers that can dynamically switch traffic to their destination, taking the advantage of multiple redundant nondedicated paths to their destination.

When service providers offer customers MPLS on their WAN links, a number of features can be implemented at the IP level by the customer instead of having to implement them at the ATM level by the carrier. These features include the ability to perform traffic engineering, support for multiple independent private data streams over a single WAN connection, implementing bandwidth reservation and Quality of Service (QoS), and more.

An example of MPLS deployment at the telco level is AT&T, which has deployed MPLS running on Cisco switches on its frame relay backbone network for more efficient transport of IP traffic over that network. MPLS is used to define different Classes of Services (CoS) and supports virtual private networking (VPN). Another example of how MPLS can be used is demonstrated by CoSine, which combines MPLS with Internet Protocol Security (IPsec) for secure tunneling and virtual routing.

MPLS is also expected to find application in pure optical backbone networks, an emerging technology driven mainly by the demand for greater capacity for Internet backbone traffic.

See Also Asynchronous Transfer Mode (ATM) ,frame relay ,Internet Protocol (IP) ,quality of service (QoS) ,routing

Multiprotocol over ATM (MPOA)

A method for forwarding Internet Protocol (IP) packets over Asynchronous Transfer Mode (ATM).

Overview

Multiprotocol over ATM (MPOA) defines a standard for forwarding Layer 3 packets (such as IP packets) over ATM backbones. MPOA is intended to support any Layer 3 network transport, but IP is the main transport used with MPOA due to its ubiquitous use in the enterprise and on the Internet. The MPOA standard is being guided by the ATM Forum.

MPOA can be used to establish connections between different IP subnets using an ATM backbone. This contracts with LAN Emulation (LANE), another IP over ATM technology (and a technology from which MPOA was derived), which is used mainly to support IP communications within a subnet. Neither MPOA nor LANE are suitable for transmission of IP traffic over wide area network (WAN) connections, however, due to the large number of ATM virtual circuits (VCs) required-nine for each pair of hosts communicating using MPOA. MPOA can also be used to connect virtual LANs (VLANs) in a switched backbone network.

Architecture

The MPOA architecture involves MPOA Servers (MPSs), also called route servers, which operate as virtual routers to switch IP packets across the ATM network. Each subnet requires one or more route servers, and IP traffic is bridged from the sending host to a route server, switched across one or more additional route servers, and then bridged directly to its destination. These route servers have intelligence that enables them to track the overall topology of the ATM network and enable IP traffic to be switched across shortcut paths between end nodes on the network. The result is low latency and delay and greater Quality of Service (QoS) for IP traffic. This architecture enables MPOA to provide the "route once, switch many" functionality needed in the core switching backbone of today's high-speed networks.

See Also Asynchronous Transfer Mode (ATM) ,Internet Protocol (IP) ,quality of service (QoS) ,routing ,virtual circuit ,virtual LAN (VLAN)

Multipurpose Internet Mail Extensions (MIME)

Extensions to Simple Mail Transport Protocol (SMTP) that allow multipart and binary messages to be sent using e-mail.

Overview

SMTP, the popular e-mail system used on the Internet, is defined in RFC 822 as a mechanism for forwarding simple text-based messages between hosts. An SMTP message consists of a single text file encoded in 7-bit ASCII format with headers attached. SMTP uses these headers to determine how the message is forwarded by SMTP hosts along the route to its destination.

Early on, two limitations of SMTP became apparent to users of the system:

Multipurpose Internet Mail Extensions (MIME) were developed to address these two concerns. MIME does not replace SMTP but simply extends it to include support for multipart message bodies and non-ASCII message content. MIME is defined in RFC 1521, and various MIME types are defined in later RFCs.

Implementation

MIME adds two additional types of headers for SMTP messages. These headers are

RFC 1521 defines seven basic message content types that can be specified in the Content-Type header of SMTP messages. Each content type can have several subtypes to further specify the type of information carried by SMTP messages. Although RFC 1521 defined some of these subtypes, succeeding RFCs have defined many more subtypes. The main MIME content types are

RFC 1521 also defines six basic data encoding methods that can be specified in the Content-Transfer-Encoding headers of SMTP messages. These are 7bit (ASCII text with line lengths less than 1000 characters), 8bit, binary, quoted-printable, base64 (Uuencoded data), and x-token. Later RFCs have defined other MIME encoding methods.

See Also e-mail ,Simple Mail Transfer Protocol (SMTP)

multiservice switch

A switch that can route packets and switch cells.

Overview

Multilevel switches are generally high-end switches that can support both Asynchronous Transfer Mode (ATM) circuit-switching and frame relay packet switching in a single chassis. Multilevel switches have been responsible in part for a rising interest in ATM backboning as a result of evolution in frame relay technologies such as Multilink Frame Relay (MFR).

Multiservice switch. A wide area network (WAN) built using multiservice switches.

In a typical wide area network (WAN) scenario, frame relay might be used at the edge of corporate local area networks (LANs) at the ingress and egress points, connecting the LANs using carrier ATM services as the internetwork backbone technology.

Multiservice switches conform to two frame relay specifications:

See Also Asynchronous Transfer Mode (ATM) , circuit-switched services ,frame relay , packet switching

Multistation Access Unit (MAU or MSAU)

A wiring concentrator (passive hub) used in Token Ring networks.

Overview

Multistation Access Units (MAUs) are the core wiring components for building Token Ring networks. MAUs direct traffic from one station to the next around the ring of a Token Ring network. While the logical topology (electrical path) of a Token Ring network is a ring, the physical topology (wiring) is actually a star topology with the MAU at the center and the stations connected in a star pattern to the MAU. The MAU thus concentrates the wiring and contains circuitry that makes the logical star network operate as a physical ring.

Implementation

A typical MAU is a stand-alone device or a rack- mounted device that provides 8 or 16 ports for connecting stations to the ring. The MAU also generally includes two additional ports: a ring-out connector and a ring-in connector. These ports are used to connect MAUs together to form larger Token Ring networks. For example, a larger network can be constructed by joining several MAUs to form a larger ring by connecting the ring-out connector of one MAU to the ring-in connector of another and continuing to connect until the ring-out of the last MAU in a series is connected to the ring-in of the first to complete the loop. Using this method, you can connect up to 33 MAUs into a large Token Ring network. Some MAUs include ring-in/out connectors for both copper cabling (RJ-45 connectors) and fiber-optic cabling (typically ST connectors).

When interconnecting MAUs to form large Token Ring networks, use MAUs that have automatic loopback functions on their ring-in/out ports. If a break in the ring should then occur between two MAUs, the break is automatically detected and traffic is rerouted along a preconfigured backup path.

MAUs often support both 4 megabits per second (Mbps) and 16 Mbps Token Ring signaling. Note, however, that a common way to bring down a Token Ring network is to connect a Token Ring station that has the wrong speed for the given network-for example, connecting a 4-Mbps station to a 16-Mbps network. This speed mismatch causes beaconing that brings down the network. A smart MAU can detect a mismatched station and lock out the station before a problem can occur.

Multistation Access Unit (MAU). Connecting several MAUs together to create a large Token Ring network.

MAU connections typically use either proprietary IBM universal data connectors for IBM Type 3 cabling (Type 1 Token Ring) or RJ-45 lobe connectors for 100-ohm shielded twisted-pair (STP) cabling (Type 3 Token Ring). MAUs also usually include circuitry that provides fault tolerance so that if a station fails or is disconnected from the MAU Token Ring, traffic can still continue unaffected around the ring. If you need to locate Token Ring stations at distances greater than 328 feet (100 meters) from a MAU, you can use stand-alone repeaters or a powered (active) MAU that has built-in repeaters to support greater lobe distances.

Other types of MAUs include

Notes

The acronym MAU also stands for Medium Access Unit or Media Attachment Unit, a term that refers to the circuitry in an Ethernet hub, Ethernet switch, or Ethernet network interface card (NIC) that enables the correct form of electrical or optical connection to be established with the particular type of media being used. This type of MAU, also known as a transceiver, detects the carrier signal and data signals on the media, notes when collisions between signals occur, and forwards this information to the remaining circuitry for processing. To distinguish the two different meanings of the acronym "MAU," the acronym MSAU is sometimes used to refer specifically to a Multistation Access Unit, as used in Token Ring networking.

See Also hub ,loopback ,shielded twisted-pair (STP) cabling ,Simple Network Management Protocol (SNMP) ,Token Ring

multitasking

Running two or more programs simultaneously.

Overview

From the point of view of the user, multitasking makes programs appear to be executing at the same time, but from the operating system's point of view, one of two things might be happening:

Types

There are two basic types of multiprocessing on machines that run versions of the Windows operating system:

Notes

Windows 2000 supports symmetric multiprocessing (SMP) on multiprocessor machines, whereby the Windows 2000 kernel lets processors share memory and assign ready threads to the next available processor. This SMP support ensures that no processor is ever idle or running a low-priority thread when a high-priority thread is waiting. Windows 2000 also supports soft affinity, whereby a thread tries to run on the same processor it last ran on, all things being equal. You can even use Task Manager to assign a specific process to a particular processor, a feature called processor affinity.

See Also asymmetric multiprocessing (AMP)

Multitenant Broadband Service Provider (MBSP)

Another name for a building-centric local exchange carrier, a telecommunications carrier focused on the Multitenant Unit (MTU) market.

See Also Building-centric Local Exchange Carrier (BLEC)

multitenant unit (MTU)

A building with many tenants, such as a skyscraper.

Overview

Multitenant units are generally large office buildings in dense urban areas, but the term can also include apartment complexes, hotels, industrial parks, and other large buildings and complexes. The MTU market is one that is rapidly growing in importance as far as telcos and other carriers are concerned-estimates indicate there are about 120,000 MTUs in the United States, and many of these MTUs host businesses that are eager for broadband data services. Traditional incumbent local exchange carriers (ILECs) have been slow to provision broadband services such as Digital Subscriber Line (DSL) for this market segment, and, as a result, many competitive local exchange carriers (CLECs) are targeting them. These carriers targeting MTUs for broadband services are generally called Multitenant Broadband Service Providers (MBSPs) or building-centric local exchange carriers (BLECs).

See Also broadband Internet access ,Building-centric Local Exchange Carrier (BLEC) ,Competitive Local Exchange Carrier (CLEC) ,Digital Subscriber Line (DSL) ,Incumbent Local Exchange Carrier (ILEC)

MUX

Stands for multiplexer, a device that combines multiple data streams into a single stream.

See Also multiplexer (MUX)

MVS

Stands for Multiple Virtual Storage, a legacy IBM mainframe operating system.

See Also Multiple Virtual Storage (MVS)

MX record

A Domain Name System (DNS) record identifying Simple Mail Transfer Protocol (SMTP) hosts.

Overview

MX records specify the DNS name and Internet Protocol (IP) address of a host that can forward e-mail for a given domain. (The acronym MX stands for mail exchange.) MX records are required for SMTP hosts to enable them to forward SMTP mail over the Internet.

Examples

An MX record specifying that incoming mail directed toward the microsoft.com domain should be forwarded to the SMTP host called mail.microsoft.com might be expressed as

microsoft.com.   IN   MX   0   mail.microsoft.com.

Here the number 0 is the preference value that specifies the priority of the SMTP host. This is used if more than one SMTP mail exchanger exists for a given domain.

See Also Domain Name System (DNS) ,resource record (RR)

My Computer

A desktop icon in Microsoft Windows that enables users to browse resources on their computer.

Overview

The My Computer icon provides a view that is different but complementary to Windows Explorer's view of system and network resources. My Computer gives users a simple way to view the resources on their computer. With Active Desktop installed, however, My Computer has an optional interface that supports browsing resources using a Web browser format.

My Computer. Typical contents of My Computer.

See Also My Documents ,My Network Places

My Documents

A folder in Microsoft Windows that gives users a convenient location for storing their documents.

Overview

My Documents is the default location where applications such as Microsoft Office save the files you create. The desktop icon for My Documents is a folder containing a sheet of paper. My Documents is also part of the user profile on a machine running Windows 2000, Windows XP, and Windows .NET Server, and you can redirect (change) the location of the target folder for My Documents by right-clicking on My Documents and selecting Properties. For example, you could modify My Documents to point to a network share where the user's documents can be safely stored and regularly backed up. Users should generally be instructed to always save their personal documents in their My Documents folder even if their applications do not automatically do so.

See Also My Computer ,My Network Places

My Network Places

A desktop icon in Microsoft Windows 2000, Windows XP, and Windows .NET Server that displays various aspects of the network your computer resides on.

Overview

My Network Places can be used to browse resources on the network. When you open My Network Places, it may contain

Note also that if you right-click My Network Places and choose Properties, you can open your Network Connections folder, which displays the types of connections that your computer has with local area networks (LANs) or the Internet and allows you to configure these connections.


Previous page
Table of content
Next page
Microsoft Encyclopedia of Networking
Microsoft Encyclopedia of Networking
ISBN: 0735613788
EAN: 2147483647
Year: 2002
Pages: 36
Authors: Mitch Tulloch, Ingrid Tulloch
BUY ON AMAZON
Beginners Guide to DarkBASIC Game Programming (Premier Press Game Development)
The Complete Cisco VPN Configuration Guide
Information Dashboard Design: The Effective Visual Communication of Data
The Oracle Hackers Handbook: Hacking and Defending Oracle
GDI+ Programming with C#
Understanding Digital Signal Processing (2nd Edition)
flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net
Privacy policy