Crossroads - Summer 2009

t f very popular in the scientific community and formed the basis of design for several supercomputers in the 1980s and 1990s. But, STAR-100, although designed to perform at 100 MFLOPS, gave lower than expected numbers in a “real-world” environment because the serialized part of the processing was still slow. Switching from vectors to normal data was still time-consuming, making the real-world performance slower than expected. This conversion theory was developed by Gene Amdahl in early 1967. However, Amdahl’s Law was ignored by architects of the STAR-100.

In 1971, Seymour Cray, unable to secure sufficient funds for his project at CDC, left the company to form Cray Research, where he designed Cray- 1 (160 MFLOPS). The Cray- 1 provided a good balance between scalar and vector performance and also used registers to dramatically improve performance. Registers are small amounts of memory storage available on processors. Their contents can be accessed at much faster speeds compared to external I/O components. However, because they reside on the processor’s chip, they are more expensive to manufacture. They also provide less flexibility in terms of size, so Cray’s machine could only read small parts of data at a time. The first release of Cray- 1 was in 1976, and it dismissed STAR-100 from its top spot as the fastest supercomputer of that time. The first official customer, the National Center for Atmospheric Research (NCAR), paid $8.86 million to own the supercomputer. This machine shaped the computer industry for years to come. Cray- 1 was also Cray’s first supercomputer to use integrated circuits (ICs).

Cray- 1 was succeeded in 1982 by Cray X-MP (800 MFLOPS), the irst multiprocessing computer, and in 1985 by Cray- 2, the first machine to break the gigaflop barrier at 1. 9 GFLOPS. Cray- 2 used all IC components instead of individual components and remained the fastest machine until 1987, when ETA Systems, a spin-off from CDC, designed a 10 GFLOPS machine called ETA- 10. ETA- 10 used fiber optics for communication between processors and I/O devices. ETA later merged back with CDC in 1989. In the meantime, two new companies, Thinking Machines Corporation (1982) and nCUBE (1983), were founded.

Both companies specialized in parallel computing architectures. Thinking Machines, started by graduates from the Massachusetts Institute of Technology, produced several supercomputers released as Connection Machines. By 1993, four of the top five fastest supercomputers belonged to Thinking Machines. nCUBE, on the other hand, was started by a group of Intel employees who wanted Intel to enter into parallel computing but couldn’t convince the decision-makers to undertake the endeavor. nCUBE released a parallel computer with the same name. In the mid-1990s, the supercomputer market collapsed, and both companies were acquired by bigger players in the business. The crash also forced Cray Research to merge with Silicon

Graphics, Inc. (SGI) in 1996.

One of the major companies that has yet to be men-ioned is IBM. Although IBM had built several of the fastest computers in the world (for example the IBM 7030), it was not until 1993 that it entered the supercomputer market with IBM SP- 1. It was the first member of the IBM’s Scalable POWERparallel distributed memory parallel computer, based on RISC System/6000 processing element, which later became known as POWER (Performance Optimization With Enhanced

RISC). In a distributed memory system, the memory and address space of each processor in a multi-processor system is local to itself. The data can only be shared between processors using a message passing interface like IBM’s message passing library (MPL). IBM continued releasing several successors to IBM-SP, and it faced stiff competition from other players in the market, such as Hitachi and Intel. At the turn of the century, IBM was at the top of the fastest supercomputer list with IBM ASCI White. It had 8,192 processors, 6 TB of memory, 160 TB of storage space, and operated at 7.226 TFLOPS.

3 2 s J

Present: Supercomputers Today

In 1993, based on ideas of Hans Meuer, a professor of Computer Science at the University of Mannheim, Germany, project TOP500 was initiated. The aim of this project was to list the 500 most powerful computer systems in the world. The list, compiled biannually, ranks supercomputers based on their performance on the LINPACK benchmark—a linear algebra library for digital computers that tests the floating point computing power of the system. Table 1 shows the list of fastest supercomputers by period since 1993.

After IBM’s ASCI White, Earth Simulator developed by NEC in apan topped the list from 2002 to 2004. It was developed to understand global climate models, and it was capable of operating at over 35 TFLOPS. IBM returned with BlueGene to reposition itself as the leader in building the fastest supercomputer. Several prototypes of BlueGene were announced: BlueGene/L (released March 2005), BlueGene/C (in development), BlueGene/P (released June 2007) and BlueGene/Q (due 2011). BlueGene remained the fastest supercomputer until 2008, when it was replaced by RoadRunner, also designed by IBM. Other powerful supercomputers released during this period include Cray’s XT- 3 Red Storm, Cray’s XT- 4 Franklin, Cray’s XT- 5 Jaguar, Dell’s Thunderbird, SGI’s Columbia, and HP’s Cluster Platform.

According to the list released in November 2008, the top three upercomputers and their average performance are:

1. IBM’s RoadRunner at Los Alamos National Laboratory, USA: 1.105

PFLOPS.

. Cray’s Jaguar XT5 at Oak Ridge National Laboratory, USA: 1.059

PFLOPS.

. SGI’s Pleiades Altix ICE 8200EX at NASA/Ames Research Center, USA: 487.01 TFLOPS.

Period /1993–11/1993

11/1993–06/1994

06/1994–11/1994

11/1994–06/1996

06/1996–11/1996

11/1996–06/1997

06/1997–11/2000

11/2000–06/2002

06/2002–11/2004

11/2004–06/2008

06/2008–06/2009

Supercomputer Name
CM- 5 (Connection Machine)
Numerical Wind Tunnel
Paragon XP/S
merical Wind Tunnel
SR 2201
CP-PACS
ASCI Red
ASCI White
Earth Simulator
BlueGene
RoadRunner