SCSI-3
| SCSI-3 is a term used to describe a set of SCSI standards currently being developed. Unlike SCSI-1 and SCSI-2, SCSI-3 is not one document that covers all the layers and interfaces of SCSI; it is instead a collection of documents that cover the primary commands, specific command sets, and electrical interfaces and protocols. The command sets include hard disk interface commands, commands for tape drives, commands for RAID, and other commands. In addition, an overall SCSI Architectural Model (SAM) exists for the physical and electrical interfaces, as does a SPI standard that controls the form of SCSI most commonly used. Each document within the standard is now a separate publication with its own revision level; for example, within SCSI-3, five versions of SPI have been published. Usually, we don't refer to SCSI-3 anymore as a specific interface and instead refer to the specific subsets of SCSI-3, such as SPI-3 (Ultra 3 SCSI). The main additions to SCSI-3 include the following:
Breaking SCSI-3 into many smaller individual standards has enabled the standard as a whole to develop more quickly. The individual standards can now be published more quickly rather than waiting for an entire large standard to be approved. Figure 7.1 shows the main parts of the SCSI architecture today. Figure 7.1. SCSI architecture. The most recent changes or additions to SCSI include the Ultra320 (Ultra 4) and Ultra640 (Ultra 5) SCSI standards, which have taken the performance of SCSI up to 320MBps and 640MBps. Ultra 2 SCSI and beyond also include the LVD electrical interface standard, which enables greater cable lengths. The older HVD signaling has also been removed from the standard. A number of people are confused about the speed variations in SCSI. Part of the problem is that speeds are quoted as either clock speeds (MHz) or transfer speeds. With 8-bit transfers, you get 1 byte per transfer, so if the clock is 40MHz (Fast-40 or Ultra 2 SCSI), the transfer speed is 40MBps. On the other hand, if you are using a Wide (16-bit) interface, the transfer speed doubles to 80MBps, even though the clock speed remains at 40MHz. With Fast-80DT, the bus speed technically remains at 40MHz; however, two transfers are made per cycle, resulting in a throughput speed of 160MBps. The same is true for Ultra 4 SCSI, which runs at 80MHz, transfers 2 bytes at a time, and has two transfers per cycle, for a throughput of 320MBps. Ultra 5 doubles the clock speed again, running at 160MHz, transferring 2 bytes at a time, two times per cycle, for a throughput of 640MBps. Finally, confusion exists because SCSI speeds or modes are often discussed using either the official termssuch as Fast-10, Fast-20, Fast-40, Fast-80DT, and Fast-160DTor the equivalent marketing terms, such as Fast, Ultra, Ultra 2, Ultra 3 (also called Ultra160), and Ultra 4 (Ultra320). Refer to Table 7.2 for a complete breakdown of SCSI official terms, marketing terms, and speeds. The further evolution of the most commonly used form of SCSI is defined under the SPI standards within SCSI-3. The SPI standards are detailed in the following sections. SPI (Ultra SCSI)The SPI standard, published in 1995, is the first SCSI standard that fell under the SCSI-3 designation. The SPI standard is officially known as ANSI/INCITS 253-1995 (formerly ANSI X3.253-1995). SPI is also called Ultra SCSI by most marketing departments, and it defines the parallel bus electrical connections and signals. A separate document, called the SCSI Interlocked Protocol (SIP), defines the parallel command set. SIP was included in the later SPI-2 and SPI-3 revisions and is no longer carried as a separate document. These are the main features added in SPI:
SPI initially included speeds up to Fast SCSI (10MHz), which enabled transfer speeds up to 20MBps using a 16-bit bus. Later, Fast-20 (20MHz), commonly known as Ultra SCSI, was added through an addendum document (ANSI/INCITS 253-1995 Amendment 1), allowing a throughput of 40MBps on a 16-bit bus (commonly called Ultra/Wide). SPI-2 (Ultra2 SCSI)SPI-2 is also called Ultra 2 SCSI and was officially published in 1998 as ANSI/INCITS 302-1998 (formerly ANSI X3.302-1998). SPI-2 adds several features to the prior versions:
The most notable of these is a higher speed, called Fast-40, which is commonly called Ultra 2 SCSI and runs at 40MHz. On a narrow (8-bit) bus, this results in 40MBps throughput, whereas on a Wide (16-bit) bus, this results in 80MBps throughput and is commonly referred to as Ultra 2/Wide. To achieve these speeds, a new electrical interface, called LVD, must be used. The slower single-ended electrical interface is good only for speeds up to Fast-20; Fast-40 mode requires LVD operation. The LVD signaling also enables longer cable lengths: up to 12 meters with multiple devices or 25 meters with only one device. LVD and SE devices can share the same cable, but in that case, the bus runs in SE mode and is restricted in length to as little as 1.5 meters in Fast-20 mode. LVD operation requires special LVD-only or LVD/SE multimode terminators. If multimode terminators are used, the same terminators work on either SE or LVD buses. The SPI-2 standard also includes SIP and defines the SCA-2 80-pin connector for hot-swappable drive arrays. There is also a new 68-pin (VHDC), which is smaller than the previous types. SCSI Signaling"Normal," or standard, SCSI uses a signaling technique called SE signaling. SE signaling is a low-cost technique, but it also has performance and noise problems. SE signaling is also called unbalanced signaling because each signal is carried on a pair of wires, usually twisted to help reduce noise. With SE, one of the pair is groundedoften to a common ground for all signalsand the other carries the actual voltage transitions. It is up to a receiver at the other end of the cable to detect the voltage transitions, which are really just changes in voltage. Unfortunately, this type of unbalanced signaling is very prone to problems with noise, electromagnetic interference, and ground leakage; these problems get worse the longer the cable is. This is why Ultra SCSI was limited to such short maximum bus lengthsas little as 1.5 meters. When SCSI was first developed, a signaling technique called HVD signaling was also introduced into the standard. Differential signaling, also known as balanced signaling, is still done with a pair of wires. In fact, the first in the pair carries the same type of signal that single-ended SCSI carries. The second in the pair, however, carries the logical inversion of that signal. The receiving device detects the difference between the pair (hence the name differential). By using the wires in a balanced pair, the receiver no longer needs to detect voltage magnitude, only the differential between voltages in two wires. This is much easier for circuits to do reliably, which makes them less susceptible to noise and enables greater cable length. Therefore, differential SCSI can be used with cable lengths of up to 25 meters, whereas single-ended SCSI is good only for 6 meters maximum, or as little as 1.5 meters in the faster modes. HVD was hardly ever used in PCs and has been replaced by LVD. A few HVD SCSI host adapters remain on the market, but HVD signaling is obsolete and was removed from SCSI-3 standards. Most LVD devices are actually multimode devices, designed to work in SE mode as well as LVD. Therefore, all multimode LVD/SE SCSI devices can be used on either LVD or SE SCSI buses. However, when on a bus with even one other SE device, all the LVD devices on the bus run only in SE mode. Because SE mode supports only SCSI speeds of up to 20MHz (Fast-20 or Ultra SCSI) and cable lengths of up to 1.5 or 3 meters, the devices also work only at that speed or lower; you also might have problems with longer cables. Although you can purchase an Ultra 3 SCSI multimode LVD/SE drive and install it on a SCSI bus along with SE devices, you will certainly be wasting the capabilities of the faster device. Fortunately, SE SCSI devices are largely obsolete, and in any case are not suitable for use in servers. Note that all Ultra 2, Ultra 3, and faster devices support LVD signaling because that is the only way they can be run at the Ultra 2 (40MHz) or Ultra 3 (80MHz) speeds. Ultra SCSI (20MHz) or slower devices can support LVD signaling, but in most cases, LVD is synonymous with Ultra 2 or Ultra 3 only. Table 7.2, earlier in this chapter, lists all the SCSI speeds and maximum lengths for each speed, using the supported signaling techniques for that speed. Because the connectors are the same for SE, HVD, LVD, or multimode SE/LVD devices, and because putting an HVD device on any bus with SE or LVD devices causes damage, it is good to be able to tell them apart. One way is to look for a special symbol on the unit; the industry has adopted different universal symbols for single-ended and differential SCSI. Figure 7.2 shows these symbols. Figure 7.2. Universal symbol icons identifying SE, LVD, multimode LVD/SE, and HVD devices. If you do not see such symbols, you can tell whether you have an HVD device by using an ohmmeter to check the resistance between pins 21 and 22 on the device connector:
Although you will blow up stuff if you plug HVD devices into LVD or SE buses, this generally should not be a problem because virtually all devices used in the PC environment are SE, LVD, or LVD/SE. HVD has essentially been rendered obsolete because it has been removed from the SCSI standard with Ultra 3 SCSI (SPI-3). SPI-3 (Ultra 3 SCSI [Ultra160])SPI-3, also known as Ultra 3 or Ultra160 SCSI, was published in 2000 and builds on the previous standard primarily by doubling the speed to Fast-80DT (double transition). This results in a maximum throughput of 160MBps. These are the main features added to SPI-3:
DT clocking sends data on both the rising and falling edges of the REQ/ACK clock. This enables Ultra 3 SCSI to transfer data at 160MBps, while still running at a bus clock rate of 40MHz. This mode is defined for 16-bit bus use only. CRC is a form of error checking incorporated into Ultra 3 SCSI. Previous versions of SCSI use simple parity checking to detect transmission errors. CRC is a much more robust form of error-detection capability that is far superior for operation at higher speeds. Domain validation allows better negotiation of SCSI transfer speeds and modes. With prior SCSI versions, when the bus is initialized, the host adapter sends an INQUIRY command at the lowest 5MHz speed to each device to determine which data transfer rate the device can use. The problem is that, even though both the host adapter and device might support a given speed, there is no guarantee that the interconnection between the devices will reliably work at that speed. If a problem occurs, the device becomes inaccessible. With domain validation, after a maximum transfer speed is negotiated between the host and device, it is then tested at that rate. If errors are detected, the rate is stepped down until the connection tests error-free. This is similar to how modems negotiate transmission speeds before communicating and really improves the flexibility and perceived reliability of SCSI. Packetization is a protocol that enables information to be transferred between SCSI devices in a much more efficient manner. Traditional parallel SCSI uses multiple bus phases to communicate different types of information between SCSI devices: one for command information, two for messages, one for status, and two for data. In contrast, packetized SCSI communicates all this information by using only two phases: one for each direction. This dramatically reduces the command and protocol overhead, especially as higher and higher speeds are used. Packetized SCSI is fully compatible with traditional parallel SCSI, which means packetized SCSI devices can reside on the same bus as traditional SCSI devices. As long as the host adapter supports packetization, it can communicate with one device using packets and another using the traditional protocol. Not all Ultra 3 or Ultra160 SCSI devices include packetization support, however. Ultra 3 devices that support packetization are typically referred to as Ultra160+ SCSI. QAS is a feature in Ultra 3 SCSI that reduces arbitration time by eliminating bus free time. QAS enables a device to transfer control of the bus to another device without an intervening BUS FREE phase. SCSI devices that support QAS report that capability in the INQUIRY command.
SPI-4, or Ultra 4 SCSI (Ultra320)SPI-4, also known as Ultra 4 or Ultra320 SCSI, was published in 2002 as ANSI/INCITS 362-2002, and it has all the same features as Ultra 3 (Ultra160) SCSI plus several new features to ensure reliable data transmission at twice the speed. Ultra320 SCSI integrates both the packetization and QAS features from Ultra160+ SCSI as mandatory features. Ultra320 SCSI then adds the following new features:
At the high 80MHz DT signaling speeds used with Ultra320, problems were discovered with high-frequency roll-off (signal degradation) and skew (timing variations) in the cables and logic boards. To compensate for this, Ultra320 can use one of two methods: write precompensation (WPC) in the transmitters or adjustable active filters (AAF) in the receivers. WPC alters the transmitted signals by driving the first bit after a clock transition harder than the subsequent bits. WPC can be turned off if receivers with AAF are used instead, because they are capable of compensating for roll-off and skew problems in the transmission line at the receiving end. With AAF, a special training pattern with low- and high-frequency signals is sent before each data transfer to allow the receiver to dynamically adjust for the high-frequency roll-off and skew problems. AAF is considered the best method because it is dynamic and automatically adjusts to compensate for the particular conditions in the bus at the time of each data transfer. Because of this, the faster Ultra640 standard requires AAF, and WPC is not an option. Ultra320 host adapters require 64-bit 133MHz PCI-X or PCI-Express x4 expansion slots, making them incompatible with very low-end servers. However, most midrange and entry-level servers include multiple PCI-X slots (which can also be used by PCI cards), and the newest designs often feature PCI-Express x4 expansion slots along with, or instead of, PCI-X slots. Although the SPI-5 (Ultra640) standard was published in 2003, it appears that Ultra320 products will be the last, and fastest, SPI-based products to be produced. SPI-5, or Ultra 5 SCSI (Ultra640)SPI-5, also known as Ultra 5 or Ultra640 SCSI, was published in 2003 as ANSI/INCITS 367-2003, and it is twice as fast as the Ultra 4 (Ultra320) standard. Its major features include the following:
With Ultra640 SCSI, ribbon cables are restricted to 2m in total length. This is because ribbon cable has high crosstalk, or signal noise transfer, between adjacent wires. The maximum length for a shielded cable bus is 10m, or 20m if a single device is connected. No products have been developed supporting SPI-5 standards. Instead, the SCSI storage industry is switching its emphasis for very high-end performance to Serial Attached SCSI (SAS). Because SAS uses the same basic physical layer and connections as SATA, SAS is becoming the next step in both SCSI and high-end SATA solutions. See the next section for details. SASWith parallel SCSI reaching the end of the road with the development of Ultra320 and the introduction of the moribund Ultra640 standard, the T10 committee decided to look into serial architecture as a possible solution for future SCSI implementations. Starting in 2002, the committee began work on a future serial SCSI standard. Rather than reinvent the wheel, it decided to adopt the same signaling, cable, and connector designs as used in SATA and combine the SCSI command protocol with a somewhat modified SATA physical layer design. The result was called SAS and was officially published in 2003 as ANSI/INCITS 376-2003. An extension featuring incremental enhancements and fixes is called SAS 1.1 and is currently under development. SAS was designed to leverage the cost economies of the SATA physical interface, cable, and connector designs while preserving the robust software and reliability of SCSI. SAS was designed as an extendable standard with future increases in speed and performance in mind. As with SATA, the serial interface is self-clocking, which allows data rates to be more easily pushed higher and higher. The main features of SAS include the following:
The most important feature of SAS is that it actually supports two types of drives. SATA drives can be attached for low-cost storage, and true SAS drives can be attached for higher-performance systems. The SAS connector design enables both SAS and SATA drives to be plugged in; however, it is impossible to plug an SAS drive in to a SATA interface. Because the SATA connector signals are a subset of SAS signals, SATA drives will operate on SAS host adapters. SAS drives, on the other hand, will not operate on SATA adapters. The SAS connectors are specially keyed to prevent plugging them in to SATA host adapters. SAS uses three protocols, each for transferring different types of data, depending on which type of device is being accessed:
By including these protocols, SAS provides full compatibility with existing SCSI devices and applications software as well as SATA devices and software. Combined with the backward compatibility of SAS and SATA physical connections, this enables SAS to operate as a universal interconnection for both SAS and SATA devices. If SAS is onboard, a user could upgrade his or her system from SATA to SAS by either replacing the SATA drive with an SAS drive or by keeping the SATA drive and adding the SAS drive to the system and using both simultaneously. To further ensure compatibility, in 2003, the SCSI Trade Association (STA) and the SATA II Working Group announced a partnership to enable system-level compatibility with SAS and SATA hard drives. SAS is designed for extremely high bandwidth. Data transfer rates start at 3GBps (300MBps), with a roadmap enabling increases up to 12GBps (1200MBps), which is nearly four times that of Ultra320 SCSI. In addition, SAS is a point-to-point connection that allows full bandwidth to each drive rather than sharing it among multiple drives, as does parallel SCSI. Devices can also use wide-port connections, meaning they can have multiple connections to a single device. The total bandwidth available in that case is multiplied by the total number of connections to the device. Internal connections use cables up to 1 meter long; these are similar to SATA cables except that they have a modified connector key (see Figure 7.3). Figure 7.3. A SATA internal hard disk drive connector compared to a SAS internal drive connector. Table 7.3 shows the pinouts of the SAS host adapter and device connectors.
External SAS connections, on the other hand, use InfiniBand-type cables that support four physical connections, along with an adapter that allows four internal connections to be mated to the external cable via an open expansion slot. The device connector looks a lot like the connector on a SATA drive; however, it also features a key and optional secondary signals for wide-port device connections. Figure 7.4 shows a typical external SAS cable and adapter. Figure 7.4. SAS external cable (left) and adapter (right). The adapter uses an empty expansion slot. The design of the host connector enables both SATA and SAS cables to be attached. The pinout of each connection uses seven pins, and the pinout is identical to the SATA pinout. Figure 7.5 shows the icon used to identify SAS devices and connections that are compliant with the SAS standard. Figure 7.5. SAS icon. A key feature of SAS is expandability. Devices called expanders enable SAS to be scaled up to connect large numbers of drives. Expanders are essentially low-cost switches that enable up to 128 individual point-to-point connections to be made; a total of up to 16,384 SAS devices can be connected via multiple expanders to a single host. In contrast, parallel SCSI imposes a limit of 15 devices per SCSI chain and severely limits total cable length. Figure 7.6 shows how expanders enable great flexibility in connections between SAS initiator devices (such as host adapters) and SAS target devices (such as SAS or SATA drives). Figure 7.6. Typical SAS expander connection diagram. Note that each expander, SAS initiator port, and SAS target port has a unique SAS address. Connections E and F represent a wide SAS initiator port with two concurrent connections to a wide SAS port. Although the SAS specification was released in October 2003, prototype SAS adapters and drives were first demonstrated by Seagate and Hewlett-Packard in March 2003. Motherboards with onboard SAS host adapters have been available since late summer 2005, and all the major disk drive manufacturers have announced plans to develop SAS disk drives; SAS allows existing SATA drives to be used as well. RAID ArraysMost servers, especially at levels above the workgroup, use SCSI drives rather than ATA drives because of their superior performance. You can enhance performance and data reliability further by creating a drive array. RAID technologies are used by both SCSI and ATA drives. Current SCSI-based RAID products primarily support Ultra320 and Ultra160 drives and are used in traditional servers and rack-mounted computers. For a complete description of RAID levels and terminologies, see Chapter 6, "The ATA/IDE Interface." Tip When you install Windows 2000, Windows XP, or Windows Server 2003 on a system with SCSI drives, make sure you press F6 when prompted to install a third-party SCSI or RAID driver. The driver must be on a floppy disk. Fibre Channel SCSIFibre Channel SCSI is a specification for a serial interface using a Fibre Channel physical and protocol characteristic with a SCSI command set. It can achieve 200MBps or 400MBps over either fiber or coaxial cable of several kilometers in length. Fibre Channel is designed for long-distance connectivity (such as several kilometers) and connecting multiple systems, and it has become a popular choice for storage area networks (SANs) and server clusters. Fibre Channel SCSI complements, rather than replaces, Ultra160 and Ultra320 SCSI, which are designed for direct connection to servers. 200MBps versions of Fibre Channel SCSI use the gigabit interface connector (GBIC), whereas 400MBps versions use either the small form factor pluggable (SFP) connector for optical connections or the high-speed serial data connector (HSSDC) for copper cable. (These connectors are shown later in this chapter, in Figures 7.117.13.) Because SAS and iSCSI offer much the same performance and benefits of Fibre Channel at a much lower cost, it is expected that SAS and iSCSI will replace Fibre Channel in many high-end server applications. iSCSIAnother variation on SCSI, iSCSI, combines the performance of SCSI drives with Ethernet networking up to gigabit speeds. Because iSCSI uses Ethernet to transport data between systems, iSCSI storage can be located anywhere an Ethernet network can reach, including Internet access. In addition, iSCSI storage enables secure remote storage for computers that could be hundreds of kilometers away. Because iSCSI data can be routed the same way any other type of Ethernet data can be routed, it enables data to be transported even when some connections between the server and the storage devices are unavailable. Eventually, iSCSI is expected to replace Fibre Channel in uses such as network attached storage (NAS), SANs, and storage clusters. Similarly to Fibre Channel, iSCSI cards can be purchased in both copper-wire and fiber-optic versions to match the Ethernet network already in use. iSCSI storage arrays are available in rack-mounted form factors for easy installation in data centers. iSCSI host adapters for copper (CAT5 and greater UTP cable) support 1Gbps (Gigabit Ethernet) transfer rates and also support Fast Ethernet (100Mbps) transfer rates. For best performance, you should use a 64-bit PCI slot, PCI-X slot or PCI-Express slot, although most are backward compatible with 32-bit PCI slots. The process of setting up an iSCSI host adapter requires the user to configure the card's parameters, identify the target drives by IP address or DNS name, specify the port number, configure login information, configure routing tables, and store configuration information in the host adapter's nonvolatile RAM. For details, see the documentation for a particular iSCSI host adapter. | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
EAN: 2147483647
Pages: 240