labels: Infotech, Intel
Tie-up of Titans: Intel and Cray form alliance to build faster supercomputers news
30 April 2008

In a significant development two industry vertical leaders - advanced supercomputer maker Cray and chipmaker Intel - have announced their decision to collaborate on developing future supercomputing technologies  (See: Cray and Intel collaborate on developing future supercomputing technologies)

To most computer users, Cray = supercomputer and Intel = personal computers (remember the line ''Intel inside''?). Those who swear by such simplistic equations would be amazed to learn that more than 70 per cent of the world's top 500 supercomputers are Intel-based, 354 to be exact. And the surprises don't end there.

Interestingly, Cray, a pioneer in this field who supplies as many as three of the top 10 supercomputer makers, does not use Intel processors in its products, relying instead on Advanced Micro Devices (AMD) or custom-built chips. All this is about to change with the latest tie-up announced between the biggest name in microprocessors and the most revered name in supercomputing.

What are supercomputers?
Before going into the intricacies of the deal, it is important to understand what supercomputers are. Basically, there is no hard-and-fast definition of a supercomputer. What is considered a supercomputer today may very well be just a fast computer next month and an ordinary one next year, such has been the blistering pace of advances in this field. For example, a modern desktop personal computer today is more powerful than a supercomputer of the 1980s.

Most tellingly, the man who is called ''the father of supercomputing'' and after whom the Cray Inc. company is named, Seymour Cray, never himself mentioned the term ''supercomputer'', although his designs had dominated the field during his life, and his company's products continue to do so even after his death.

A supercomputer can simply be a computer that is considered, or was considered at the time of its introduction, to be at the frontline in terms of processing capacity, particularly speed of calculation. The term ''super computing'' was first used by the New York World newspaper in 1929 to refer to large custom-built tabulators that IBM had made for Columbia University.

The changing face of supercomputing
Supercomputers introduced in the 1960s were designed primarily by Seymour Cray at Control Data Corporation (CDC), and led the market into the 1970s until Cray left to form his own company, Cray Research in 1972, whose current avatar is Cray Inc.

At his company, Cray held the top spot in the field till 1990, before traditional companies like International Business Machines (IBM) and Hewlett-Packard (HP) entered the fray in a big way. Cray's iconic product, the Cray-2, was the world's fastest supercomputer from 1985 to 1989, a very long period considering the rapidity with which the first position of the supercomputing changes names nowadays.

The typical structure of a supercomputer has also grown through a radical change over the years. CDC's early machines were simply very fast scalar processors, some 10 times the speed of the fastest machines offered by other companies. In the 1970s most supercomputers were dedicated to running a vector processor, and many of the newer players developed their own such processors at a lower price to enter the market.

Scalar processors – Processes only one data item at a time.
Vector processors – Single instruction operates simultaneously on multiple data items.

Then, with the 1980s came machines with a modest number of vector processors working in parallel. In the later 1980s and 1990s, attention turned from vector processors to massive parallel processing systems with thousands of ''ordinary'' CPUs, some being off the shelf units and others being custom designs.

Nowadays, parallel designs are based on ''off the shelf'' server-class microprocessors, such as the AIM PowerPC, Intel's Itanium, or its and AMD's x86-64, and most modern supercomputers are now highly-tuned computer clusters using commodity processors combined with custom interconnects.

PowerPC is a microprocessor architecture created by the 1991 Apple–IBM–Motorola alliance, known as AIM. Originally intended for personal computers, PowerPC CPUs have since become popular embedded and high-performance processors as well. It is used in Sony's Playstation 3, and available in both 32-bit and 64-bit variants.

Itanium is the brand name for 64-bit Intel microprocessors that implement the Intel Itanium architecture (formerly called IA-64). It is the fourth most popular architecture in large enterprise systems.

x86-64 is a 64-bit superset of the x86 instruction set architecture. It was designed by AMD, who have since renamed it AMD64. Intel has cloned it under the name Intel 64. This is the most popular choice for supercomputers.

Supercomputers predominantly run some variant of Linux or UNIX. Linux has been the most popular operating system since 2004.

How are supercomputers compared?
The speed of a supercomputer is generally measured in "FLOPS" (fLoating point operations per second), commonly used with an SI prefix such as tera-, combined into the shorthand ''TFLOPS'' (1012 FLOPS, pronounced teraflops), or peta-, combined into the shorthand ''PFLOPS'' (1015 FLOPS, pronounced petaflops).

As of today, the IBM Blue Gene / L at Lawrence Livermore National Laboratory (LLNL) is the fastest operational supercomputer, with a sustained processing rate of 478.2 TFLOPS. However, its position may be at risk post this Intel-Cray deal, because one of its stated objectives is to build the world's first PFLOF supercomputer, or one capable of performing quadrillions of machine instructions a second at double the speed of the Blue Gene.

An interesting fact to be proud of: India's EKA (the Sanskrit name for number one) supercomputer is ranked as the 4th fastest in the world and is the fastest supercomputer in Asia, according to the Top 500 Supercomputer list announced at SC07, the International Conference for High Performance Computing, Networking, Storage and Analysis at Reno, Nevada, USA.

Intricacies of the deal
The two companies plan to explore future supercomputer component designs such as multi-core processing and advanced interconnects. As a result of this collaboration, Cray and Intel plan to develop a range of high-performance computing (HPC) systems and technologies over the next several years. However, Cray would like to hedge its bets and continue using AMD chips.

Cray aims to deliver its first PFLOP-scale machine later in 2008 at Oak Ridge National Laboratory. The Intel tie-up will help Cray fulfill its product technology roadmap code-named "Cascade," which the company plans to deliver by 2011 or 2012, according to Ian Miller, Cray's head of sales and marketing. ''Cascade'' is being developed under grant from the US Defense Advanced Research Projects Agency (DARPA).

With the new Nehalem microarchitecture due out later in 2008, Kirk Skaugen, vice president of Intel's Digital Enterprise Group, said Cray and other vendors can take advantage of new technologies such as an integrated memory controller and what Intel called the "QuickPath Interconnect," which replaces the front side bus and creates a high-speed, low-latency way to connect chips together.

The next stage in supercomputing
When it incorporates the QuickPath Interconnect into its own systems, Cray will be able to connect its vector processors, which allow for high-memory bandwidth, to Intel's standard x86 processors, which should increase the overall performance of these supercomputers.

Skaugen even predicted the possibility of reaching the next stage in supercomputer evolution with this new agreement, by building an EFLOP or exaflop computer, a thousand times faster than a PFLOP machine.

A year-and-a-half ago, Santa Clara, California-based Intel formed a new business division Digital Enterprise Group to target the high performance computer market, one of the fastest-growing segments in the computer server industry. The HPC market will grow to $16 billion by 2012 from $10 billion in 2006, executives from the companies said, citing data from market research firm IDC.

Response of the management
The management teams of both companies were visibly excited at this tie-up that brings together two of the biggest names in the computing world.

Peter Ungaro,president and CEO, Cray Inc''We're excited at the potential of bringing together Intel's powerful silicon expertise and Cray's industry leadership in scalable HPC systems,'' said Peter Ungaro, president and CEO of Cray. ''We pride ourselves in offering the most innovative supercomputing systems and our customers will now enjoy greater choice in processor technologies.''

"This collaboration provides the HPC market segment with access to the best microprocessors the industry has to offer at any point in time, in the most advanced supercomputers in the world," he added. "This further strengthens Cray's industry-leading adaptive supercomputing vision as we move into the Cascade timeframe and beyond."

"Cray's commitment to Intel is a testament of our commitment to HPC and the strength of our hardware and software roadmap and many-core research," said Patrick Gelsinger, senior vice president and general manager of Intel's Digital Enterprise Group.

''Throughout Cray's history, it has been an innovator in high-end HPC while Intel has pushed the boundaries of processor technology. The combination of this industry leadership and technical strength will allow HPC users to take advantage of future Xeon and other Intel processor technologies,'' he added. ''Together we will enable fundamental and historical problems of science and industry to be solved.''


 search domain-b
  go
 
Tie-up of Titans: Intel and Cray form alliance to build faster supercomputers