Increase performance with RAID

A robust RAID solution attached to the new Ultra2 SCSI interface holds out the promise of better performance and reliability

A robust RAID solution attached to the new Ultra2 SCSI interface holds out the promise of better performance and reliability

Applications such as digital video, transaction processing and data warehousing have placed unprecedented requirements upon computers and peripheral I/O channels. Distributed data across wide area and local area networks have created an exponential demand for servers that can handle small packet sizes and a high number of I/O requests.

All of the above has pushed server designers to move to RAID solutions which are truly scaleable in I/O performance. To meet part of the challenge, a new high bandwidth interface, called Ultra-2 SCSI (LVD), is entering the market. It allows the server designer to remove the peripheral channel I/O. Adding to that, new performance RAID controllers are available which solve the I/O scalability problem.

SCSI was created to satisfy the need for a more flexible, faster, command-controlled interface for hard disk drives and other computer peripherals. Despite the term "small" in its name, SCSI (Small Computer System Interface) is large. It is large in use, market impact, influence, and unfortunately, documentation. Early SCSI was simple, it had a few dozen commands and the specification could be tucked in your pocket. The standards effort that began with a 20-page specification has grown to a 600-page extravaganza of technical information. In 1985, a group of manufacturers approached the X3T9.2 Task Group when the first SCSI standard was being finalised as an ANSI standard.

The group wanted to increase the mandatory requirements of SCSI and define further features for direct-access devices. Rather than delay the SCSI standard, the Task Group formed an ad hoc group to develop a working paper that was eventually called the Common Command Set (CCS). Many products were designed to this working paper. The standard provided what were, at the time, marvelous new performance characteristics - it could connect a host with as many as seven different devices and data could be moved at an exhilarating 5Mb/s.

Even before ANSI published the first SCSI standards document in 1986, the ASC (Accredited Standards Committee) X3T9.2 was hard at work on SCSI-2. While SCSI-2 was to go beyond the original SCSI standard (now referred to as SCSI-1), it was to retain a high degree of compatibility with SCSI-1 devices. No technical rationale can be offered as to where SCSI-1 ended and SCSI-2 began, or as to where SCSI-2 ended and SCSI-3 began. The justification is much simpler - you have to stop sometime and get a standard printed. Popular interfaces never stop evolving, adapting and expanding to meet more uses than originally envisaged.

An ongoing standardisation effort began to extend the capabilities of SCSI-2. Several options were removed from SCSI-1 as the document moved to SCSI-2. New low-level requirements in SCSI-2 along with many options significantly enhanced SCSI. Advanced command sets were added to SCSI-2 and command sets were improved. Optional messages were added to negotiate wide transfers to support command queuing which extends up to 256 commands instead of one command. Sense keys and sense codes were formalised and extended.

More importantly, SCSI was broken up from a single document to different documents which separated out the physical and logical layers. This was done to allow for different physical transport layers such as Fibre Channel and SSA to utilise SCSI's rich command set.

Ultra-2 SCSI is the next major performance advancement to the Small Computer System Interface and is a lower cost alternative to serial SCSI (Fibre Channel. Ultra-2 SCSI boosts data transfer rates from the current limit of 20Mb/s to 40Mb/s. It also doubles the Fast Wide SCSI- 2 data transfer rate from 40 to 80Mb/s.

These performance improvements create the bandwidth necessary to support the data intensive applications to be used in the coming generations of video, publishing and high-end servers. The increased data transfer rates are attributed to the faster cycle times for data transfer and the arbitration of SCSI commands. Host systems and devices using Ultra-2 SCSI will be able to negotiate optimal parameters for speed, width, offset and so on. These improvements and cycle times are primarily fueled by the higher speed of the new semiconductor technologies employed in SCSI chipsets.

Strategically, Ultra-2 SCSI is the logical migration from Ultra SCSI to accommodate optimal system performance and ever-increasing demands by users for higher performance. While the infrastructure to support Fibre Channel, SSA and other serial interfaces are being developed, Ultra SCSI is the cost-effective solution for servers and workstations in the near term. By effectively extending the life of the standard parallel interface, SCSI provides a much-needed stepping stone to the serial interface of the future.

Ultra-2 SCSI is backward compatible and uses the same physical environment as Ultra SCSI. Ultra-2 SCSI drives and systems can operate at the lower speeds of earlier SCSI versions for compatibility with older hardware. Cables connectors and terminators that support SCSI can support Ultra-2 SCSI. More importantly, Ultra-2 SCSI can be integrated without having to modify or change operating systems.

Finally, new generations of high-speed multiprocessor servers are emerging. In today's disk drives areal densities are increasing with hard disk drive manufacturers rolling out drives with faster I/Os/sec and higher sustained data transfer rates. The net result has been a growing demand for higher performance disk servers, which means a requirement for Ultra-2 SCSI.

Drive manufacturers are working closely with host adaptor companies to ensure compatibility across the SCSI bus and include: wide SCSI with data transfer at bus widths of 16 and 32-bits; fast SCSI which achieves 10Mb/s transfer rate with wider data paths of 16-bits, rising bandwidth up to 20Mb/s; and active termination.

Higher I/O requirements at both the drive side and the host side of the SCSI bus absolutely requires a leap in peripheral I/O technology. Technologies such as Fibre Channel have shown bandwidth to accommodate much higher data rates than that ( up to 1Gb/s. However, the transition to a new serial drive interface won't be as fast, smooth or inexpensive.

Interoperability and compatibility issues for systems manufacturers have emerged. This forces many system manufacturers to postpone their decision rather than lock themselves into a potentially obsolete technology. And therein lies the catch: System OEMs need a high I/O peripheral solution now, not two or three years from now.

Fortunately, such a solution has arrived in the form of the new Ultra-2 SCSI (LVD) interface as proposed by the ANSI XT310 committee. Ultra-2 SCSI (LVD) represents an enhancement of the current parallel SCSI interface standard. It has the following benefits:

It doubles SCSI drives' burst data transfer rates to 80Mb/s

It allows up to 15 devices to be connected

Some of the benefits of Ultra-2 SCSI are easy integration, reusability of current hardware and firmware, reusability of current test equipment and lower inventory risk. Due to Ultra-2 SCSI's backward compatibility, it is a low-cost solution to higher performance. The first Ultra-2 SCSI drives are already on the market.

Obtaining faster data transfers can pose challenges with analog chip design, which consists of I/O driver cells or transceivers. The inherent difficulty with a SCSI interface is that the cable, which links the drive and the host system, is a transmission line connection. The fast signal transitions, or skew rates, associated with high-performance analog circuit designs often create transmission-line effects such as ringing, overshoot and undershoot which cause signal quality problems and can interfere with the interface chip's transceiver performance.

The traditional solution is to slow down skew rates. This is because it would also prevent the interface chip from meeting the timing margins needed for Ultra-2 SCSI data transfer speeds. Instead, Ultra-2 SCSI's developers drew upon their experience with potential transmission-line effects: By using "Low Voltage Differential" technology the effective length of the Ultra-2 SCSI bus is now 25m for two devices or 1m for 15 devices. This technology allows older devices to be connected at a slower bus speed to maintain downward compatibility.

The proliferation of disk drives in servers, RAID and video systems can stress SCSI's logical connectivity. Under the existing SCSI arbitration protocol, the logical number of devices that can be addressed is limited to the number of available data lines: eight devices including the host for narrow SCSI, 16 devices for wide SCSI.

But there is a growing need for more connections. One means of addressing this need is a two-phase arbitration technique called dual-phase protocol, currently under development in ANSI's SCSI committee. The result of the first phase of addressing determines which devices are eligible to participate in the second round. The lower bits are then used to determine the selected device. With this scheme, the number of addressable devices is equal to the square of the number of data lines or up to 256 devices with a 16-bit interface. Or, for those who demand the ultimate limit, up to 1,024 devices with a 32-bit parallel interface.

Pushing connectivity into the Storage Area Network, Ultra-2 SCSI is leading the technology trend. Fibre Channel to Ultra-2 SCSI RAID bridges are available today. This allows multiple arrays of drives to be attached to an external controller. In some configurations, as many as four Ultra-2 SCSI channels are supported.

The embedded RAID subsystem is the engine behind server performance. It typically includes thousands of lines of RAID software from which a modern server is operated. It provides many storage-related functions including:

Data I/O

Remote Management Across The Network

Remote Configuration Across The Network

Support for RAID levels

Error Alerts

Other Standard RAID features

To support these features while running a large number of I/O processes requires a more powerful DSP. As I/O requests are processed, the DSP can service them more effectively. Additionally, as drives are added to the Ultra-2 SCSI bus, a less powerful RAID controller may saturate and not scale properly. The net result is that either another controller is added to scale the servers I/O processing power or the server reaches a "glass ceiling".

With a scaleable RAID controller, the server starts to saturate as the numbers of drives are increased. To provide scalability, two key elements are needed, a modern high speed DSP and a 64 bit PCI interface.

Compiled by Ajith Ram

( 1998 Mylex UK

W.DAD.WP4-T1.190799.DOC I.S. Department 19/08/99 11:40

Read more on Integration software and middleware