Cray

Cray Inc.
Type Public
Traded as NASDAQCRAY
Founded 1972 as Cray Research, Inc.
Founder(s) Seymour Cray
Headquarters Seattle, Washington, United States
Key people Peter J. Ungaro
Products Supercomputers
Employees 800[1]
Website cray.com

Cray Inc. (NASDAQCRAY) is an American supercomputer manufacturer based in Seattle, Washington. The company's predecessor, Cray Research, Inc. (CRI), was founded in 1972 by computer designer Seymour Cray. Seymour Cray went on to form the spin-off Cray Computer Corporation (CCC), in 1989, which went bankrupt in 1995, while Cray Research was bought by SGI the next year. Cray Inc. was formed in 2000 when Tera Computer Company purchased the Cray Research Inc. business from SGI and adopted the name of its acquisition.

Contents

Company history

Background: 1950 to 1972

Seymour Cray began working in the computing field in 1950 when he joined Engineering Research Associates (ERA) in Saint Paul, Minnesota. There, he helped to create the ERA 1103, generally regarded as the first successful scientific computer. ERA eventually became part of UNIVAC, and began to be phased out. He left the company in 1960, a few years after former ERA employees set up Control Data Corporation (CDC). He eventually set up a lab at his home in Chippewa Falls, Wisconsin, about 85 miles to the east. Cray had a string of successes at CDC, including the CDC 6600 and CDC 7600.

Cray Research Inc. and Cray Computer Corporation: 1972 to 1996

When CDC ran into financial difficulties in the late 1960s, development funds for his follow-on CDC 8600 became scarce. When he was told the project would have to be put "on hold" in 1972, Cray left to form his own company, Cray Research Inc., with research and development facilities in Chippewa Falls and the business headquarters in Minneapolis. The company's first product, the Cray-1 supercomputer was a major success when it was released. It was faster than all other computers at the time. The first system was sold within a month for US$8.8 million. Seymour Cray continued working, this time on the Cray-2, though it only ended up being marginally faster than the Cray X-MP, developed by another team at the company.

He soon left the CEO position to become an independent contractor. Cray started a new VLSI technology lab for the Cray-2 in Boulder, Colorado, Cray Laboratories, in 1979. The Labs were closed in 1982, but Cray later headed a similar spin-off in 1989, forming Cray Computer Corporation (CCC) in Colorado Springs. Seymour Cray worked there on the Cray-3 project, the first attempt at major use of gallium arsenide (GaAs) semiconductors in computing. However, the changing political climate (collapse of Warsaw Pact and the end of Cold War) resulted in poor sales (only one Cray-3 was delivered), and the company fell by the wayside, eventually filing for bankruptcy in 1995. CCC's remains then began Cray's final corporation SRC Computers, Inc. which still exists.

Cray Research continued development along a separate line of computers, originally with lead designer Steve Chen and the Cray X-MP. After Chen's departure, the Cray Y-MP, Cray C90 and Cray T90 were developed. All were based on the original Cray-1 architecture, but added multiple processors, faster clocks and wider vector pipes to achieve much greater performance. Because of the uncertainty of the Cray-2 project, a number of Cray-object-code compatible "Crayette" firms started: Scientific Computer Systems (SCS), American Supercomputer, Supertek, and perhaps at least one other firm. Not meant to compete against Cray, these firms attempted less expensive, slower CMOS versions of the X-MP with the release of the COS operating system (SCS) and the CFT Fortran compiler. All these firms also considered National labs (LANL/LLNL) developed CTSS operating system as well before caving in to the tide of Unixes.

In the late 1980s the high-performance market began to be overtaken by a series of massively parallel computers, led by pioneers Thinking Machines, Kendall Square Research, Intel Supercomputing Systems Division, nCUBE, MasPar and Meiko Scientific. At first, Cray Research denigrated such approaches, complaining that developing software to effectively use the machines was difficult — which was true in the era of the ILLIAC IV, but becoming less so each day. Eventually Cray realized the approach was likely the only way forward and started a five year project to capture the lead in this area as well. The result was the DEC Alpha-based Cray T3D and Cray T3E series, which left Cray as the only remaining supercomputer vendor in the market besides NEC, by 2000.

Cray computers were extremely expensive machines, and the number of units sold was small compared to ordinary mainframes. Thus, most sites with a Cray installation considered it quite prestigious to be a member of the "exclusive club" of Cray operators. This perception extended to countries as well. To boost the perception of exclusivity, Cray Research's marketing department had promotional neckties made with a mosaic of tiny national flags illustrating the "club of Cray-operating countries".(Computer History Museum, Cray 1 30th Anniversary recorded presentation, 2006)

In the late 1980s and early 1990s a number of new vendors introduced small supercomputers, known as minisupercomputers (as opposed to superminis), which started to erode the market that might have otherwise considered a low-end Cray machine. Particularly popular was the Convex Computer series, as well as a number of small-scale parallel machines from companies like Pyramid Technology and Alliant Computer Systems. One such company was Supertek, whose S-1 machine was an air-cooled CMOS implementation of the X-MP processor. Cray purchased Supertek in 1990 and sold the S-1 as the Cray XMS, but the machine proved problematic. Meanwhile their not-yet-completed S-2, a Y-MP clone, was later offered as the Cray Y-MP EL (later becoming the EL90 series), which started to sell in reasonable numbers in 1991-92. These systems were sold to smaller companies, notably in the oil exploration business. This line evolved into the Cray J90 and eventually the Cray SV1 in 1998.

In December 1991, Cray purchased some of the assets of Floating Point Systems, another minisuper vendor who had moved into the file server market with their SPARC-based Model 500 line.[2] These SMP machines scaled up to 64 processors and ran a modified version of Sun Microsystems' Solaris. Cray set up Cray Research Superservers, Inc. (later the Business Systems Division) to sell this system as the Cray S-MP, later replacing it with the Cray CS6400. In spite of these machines being some of the most powerful available when applied to appropriate workloads, Cray was never very successful in this market, possibly due to it being so foreign to their existing market niche.

Silicon Graphics: 1996 to 2000

Cray Research merged with Silicon Graphics (SGI) in February 1996. At the time the industry was highly critical of the move, noting that there was little overlap between the two companies in terms of market or technology. Founder Seymour Cray died as a result of a traffic accident later that year.

SGI immediately sold off the Superservers business to Sun, who quickly turned the UltraSPARC-based Starfire project then under development into the extremely successful Enterprise 10000 range of servers.

SGI did use a number of Cray technologies in their attempt to move from the graphics workstation market into supercomputing. Key among these was the use of the Cray-developed HIPPI data-bus and details of the interconnects used in the T3 series.

SGI's long-term strategy was to merge their high-end server line with Cray's product lines in two phases, code-named SN1 and SN2 (SN standing for "Scalable Node"). The SN1 was intended to replace the T3E and SGI Origin 2000 systems and later became the SN-MIPS or SGI Origin 3000 architecture. The SN2 was originally intended to unify all high-end/supercomputer product lines including the T90 into a single architecture. This goal was never achieved before SGI divested itself of the Cray business, and the SN2 name was later associated with the SN-IA or SGI Altix 3000 architecture.

Under SGI ownership, one new Cray model line, the SV1, was launched in 1998. This was a clustered SMP vector processor architecture, developed from J90 technology.

SGI set up a separate Cray Research Business Unit in August 1999 in preparation for detachment. On March 2, 2000, the unit was sold to Tera Computer Company. Tera Computer Company was then renamed Cray Inc. when the deal closed on April 4.

Cray Inc.: 2000 to present

After the Tera merger, the Tera MTA system was relaunched as the Cray MTA-2. This was not a commercial success and shipped to only two customers. Cray Inc. also badged the NEC SX-6 supercomputer as the Cray SX-6 and acquired exclusive rights to sell the SX-6 in the USA, Canada and Mexico.

In 2002, Cray Inc. announced their first new model, the Cray X1 combined architecture vector / MPP supercomputer. Previously known as the SV2, the X1 is the end result of the earlier SN2 concept originated during the SGI years. In May 2004, Cray was announced to be one of the partners in the U.S. Department of Energy's fastest-computer-in-the-world project to build a 50 teraflops machine for the Oak Ridge National Laboratory. As of November 2004, the Cray X1 had a maximum measured performance of 5.9 teraflops, being the 29th fastest supercomputer in the world. Since then the X1 has been superseded by the X1E, with faster dual-core processors.

On October 4, 2004, the company announced the Cray XD1 range of entry-level supercomputers which use dual-core 64-bit AMD Opteron CPUs running Linux. This system was previously known as the OctigaBay 12K before Cray's acquisition of that company. The XD1 provides one Xilinx Virtex II Pro field-programmable gate array (FPGA) with each node of four Opteron processors. The FPGAs can be configured to embody various digital hardware designs and so can augment the processing or input/output capabilities of the Opteron processors. Furthermore, each FPGA contains a pair of PowerPC 405 processors which can add to the already considerable power of a single node.

In 2004, Cray completed the Red Storm system for Sandia National Laboratories. Red Storm has processors clustered in 96 unit cabinets, a theoretical maximum of 300 cabinets in a machine, and a design speed of 41.5 teraflops. The Cray XT3 massively parallel supercomputer is a commercialized version of Red Storm, similar in many respects to the earlier T3E architecture, but, like the XD1, using AMD Opteron processors. The Cray XT4, introduced in 2006 added support for DDR2 memory, newer dual-core and future quad-core Opteron processors. The XT4 also allowed FPGA chips to be plugged directly into processor sockets, unlike the XD1, which required a dedicated socket for the FPGA coprocessor. The XT4 also used the second generation SeaStar2 communication coprocessor. [1]

On November 13, 2006, Cray announced a new system, the Cray XMT, based on the MTA series of machines. [2] [3]. This system combines multi-threaded processors, as used on the original Tera systems, and the SeaStar2 interconnect used by the XT4. By reusing ASICs, boards, cabinets, and system software used by the comparatively higher volume XT4 product, the cost of making the very specialized MTA system can be reduced. A second generation of the XMT is scheduled for release in 2011, with the first system ordered by the Swiss National Supercomputing Center (CSCS).[3]

In 2006, Cray announced a vision of products dubbed 'Adaptive Supercomputing'.[4] The first generation of such systems, dubbed the Rainier Project, used a common interconnect network, programming environment, cabinet design, and I/O subsystem. These systems included the XT4 and the XMT. The second generation, launched as the XT5h, allowed a system to combine compute elements of various types into a common system, sharing infrastructure. The XT5h combined Opteron, vector, multithreaded, and FPGA compute processors in a single system. Next generation Cascade[5] systems will make use of future multicore and/or manycore Opteron and Intel processors as well as vectorization and multithreading accelerators. Cascade is scheduled to be introduced sometime between 2011 and 2013.

In April 2008, Cray and Intel announced they would collaborate on future supercomputer systems. This partnership produced the Cray CX1 system, launched in September the same year. This is a blade server system, comprising up to 16 dual- or quad-core Intel Xeon processors, with either Microsoft Windows HPC Server 2008 or Red Hat Enterprise Linux installed.

In early 2010, Cray introduced the Cray CX1000, a rack-mounted system with a choice of compute-based, GPU-based, or SMP-based chassis. In May 2010 the Cray XE6 supercomputer was announced. The Cray XE6 system, built on Cray XT5 technology, combines the Gemini system interconnect with AMD multicore processors. The first multi-cabinet XE6 system was shipped in July 2010.

As of May 2011, the largest computer system Cray has delivered is the XT5 system at National Center for Computational Sciences at Oak Ridge National Laboratories.[6] This system, with over 224,000 processing cores, is dubbed "Jaguar" and was the fastest computer in the world as measured by the LINPACK benchmark[7] at the speed of 1.75 petaflops[8] until being surpassed by the Tianhe-1A in October 2010. It is the fastest system available for open science and was the first system to exceed a sustained performance of 1 petaflops on a 64-bit scientific application; it has since been upgraded to a peak performance of more than two petaflops.

On May 24, 2011, Cray announced the Cray XK6 hybrid supercomputer. The Cray XK6 system, capable of scaling to 500,000 processors and 50 petaflops of peak performance, combines Cray's Gemini interconnect, AMD's multi-core scalar processors, and NVIDIA's many-core GPGPU processors.

References

Further reading

External links