Glossary

---

ASM

Advanced Systems Management processor (also known as service processor) allows remote node power on/off/reset capability, monitors node environmental conditions, and allows remote POST/BIOS console, power management, and SNMP alerts.

Beowulf

An approach to building a cluster of off-the-shelf commodity personal computers, interconnecting them with Ethernet, and running programs written for parallel processing.

Bootstrap Protocol (BOOTP)

A protocol that allows a client to find both its Internet Protocol (IP) address and the name of a file from a network server.

cluster

A loosely-coupled collection of independent systems (nodes) organized into a network for the purpose of sharing resources and communicating with each other.

Cluster Systems Management (CSM)

A distributed system management solution for machines or nodes. With this software, an administrator can easily set up and maintain a system cluster by using functions like monitoring, hardware control, and configuration file management. Specifically, within the cluster, nodes can be added, removed, changed, or listed. Commands can be run across nodes or node groups in the cluster, and responses can be gathered. Nodes and applications can be monitored as to whether they are up or down, CPU, memory, and system utilization can be monitored, and automated responses can be run when events occur in the cluster. A single management server is the control point for the CSM cluster. Note that CSM manages a loose cluster of machines. It does not provide high availability services or fail-over technology, although high-availability clusters can be part of the set of machines that CSM is managing.

disk descriptor

Defines how a disk is to be used within a GPFS file system. This data is stored on the primary and secondary GPFS cluster data servers as defined in GPFS commands.

Domain Name System (DNS)

A server program that supplies name-to-address conversion by mapping domain names to IP addresses. The domain name server allows users to request services of a computer by using a symbolic name, which is easier to remember than an IP address.

Dynamic Host Configuration Protocol (DHCP)

An application-layer protocol that allows a machine on the network, the client, to get an IP address and other configuration parameters from the server.

High Availability (HA)clusters

A system typically made up of two or more robust machines that mirror each other. Usually half the nodes are actively working, while the other half are quietly waiting to take over in case of a failure. HA cluster design varies according to its function.

High Performance Computing cluster (HPC cluster)

A system designed for greater computational power than a single computer alone could provide. In the past, high-performance computing was typically reserved for the scientific community, but now it is breaking into the mainstream market. HPC is a field of computer science that focuses on developing supercomputers, parallel processing algorithms, and applications.

General Parallel File System (GPFS)

Allows users shared access to files that may span multiple disk drives on multiple nodes. It offers many of the standard UNIX file system interfaces, allowing most applications to execute without modification or recompiling. GPFS also supports UNIX file system utilities, that is, users can continue to use the UNIX commands they have always used for ordinary file operations. The only commands unique to GFPS are those for administering the GPFS file system. GPFS provides file system services to parallel and serial applications. GPFS allows parallel applications simultaneous access to the same files, or different files, from any node in the GPFS nodeset, while managing a high level of control over all file system operations. GPFS is particularly appropriate in an environment where the aggregate peak need for data exceeds the capability of a distributed file system server.

GPFS cluster

A subset of existing cluster nodes defined as being available for use by GPFS file systems. The GPFS cluster is created via GPFS commands. GPFS nodesets and file systems are subsequently created by other GPFS commands.

GPFS nodeset

A GPFS nodeset is a group of nodes that all run the same level of GPFS code and operate on the same file systems. Users have the ability to define more than one GPFS nodeset in the same GPFS cluster.

Inter-Process Communication (IPC)

The process by which programs communicate data to each other and synchronize their activities. Semaphores, signals, and internal message queues are common methods of interprocess communication.

KickStart

Allows users to automate most of a Red Hat Linux installation. A system administrator can create a single file containing the answers to all the questions that would normally be asked during a typical Red Hat Linux installation.

Message Passing Interface (MPI)

Message-passing standard that facilitates the development of parallel applications and libraries.

metadata

Data structures that contain access information about file data. These might include inodes, indirect blocks, and directories. These data structures are used by GPFS but are not accessible to user applications.

mirroring

The creation of a mirror image of data to be reserved in the event of disk failure.

Network File System (NFS)

A distributed file system that allows users to access files and directories located on remote computers and to treat those files and directories as if they were local.

Network Shared Disk (NSD)

A GPFS function that allows application programs executing at different nodes of a GPFS cluster to access a raw logical volume as if it were local at each of the nodes. In actuality, the logical volume is local at only one of the nodes, known as the server node.

open source

Software that is developed and improved by a group of volunteers cooperating together on a network. Many parts of the UNIX operating system were developed this way, including Linux.

Preboot Execution Environment (PXE)

Designed to define a standard set of preboot protocol services within a client, with the goal of allowing network-based booting to use industry-standard protocols.

PVM

Portable message-passing programming system designed to link separate host machines to form a "virtual machine," which is a single, manageable computing resource.

PXELinux

A SysLinux derivative for booting Linux off a network server using a network ROM conforming to PXE specifications.

RSA

Remote Supervisor Adapter (also known as service processor) allows remote management similar to ASM. However, RSA provides additional functions over the ASM adapters. RSA is used with the new models like the Model 342.

Red Hat Package Manager (RPM)

A packaging system that allows users to package source code for new software into source and binary form such that binaries can be easily installed and tracked and source can be rebuilt easily.

Redundant Array of Independent Disks (RAID)

A set of physical disks that act as a single physical volume and use parity checking to protect against disk failure.

Serial Storage Architecture (SSA)

An expanded storage adapter for multi-processor data sharing in UNIX-based computing, allowing disk connection in a high-speed loop.

Simple Network Management Protocol (SNMP)

A protocol governing network management and the monitoring of network devices and their functions.

small computer system interface (SCSI)

An adapter supporting the attachment of various direct-access storage devices.

storage area network (SAN)

The connectivity of multiple systems that attach to a single storage device.

SysLinux

A boot loader for the Linux operating system that promises to simplify the Linux installation.

System Resource Controller (SRC)

A set of commands and subroutines to make it easier for the system manager and programmer to create and control subsystems. A subsystem is any program or process or set of programs or processes that is usually capable of operating independently or with a controlling system.

Trivial File Transfer Protocol (TFTP)

A set of conventions for transferring files between hosts using minimal protocol.

virtual file system (VFS)

An abstraction of a physical file system implementation that provides a consistent interface to multiple file systems, both local and remote. This consistent interface allows the user to view the directory tree on the running system as a single entity even when the tree is made up of a number of diverse file system types.

VLAN

Short for virtual LAN. A network of nodes that behave as if they were connected to the same cable even though they may actually be physically located on different segments of a LAN. VLANs are configured through software rather than hardware, which makes them extremely flexible.

virtual shared disk (VSD)

A virtual component where nodes have access to remotely-mounted raw logical volumes.

---