INTEL TO BUILD CHIPS FOR INFINIBAND

August 25, 2000

COMMERCIAL NEWS

San Jose, CA — Stephen Shankland reports that InfiniBand, a technology for high-speed communication among servers, storage systems and networks, has taken an important step closer to reality.

Joining other chipmakers embracing the technology, Intel announced this week that it will build chips for InfiniBand, a standard that an alliance of industry giants is currently developing.

Although Intel’s support of InfiniBand was not in doubt, the chipmaker announced plans that bring the technology closer to fruition. Intel will make all three types of chips needed for InfiniBand systems and will license parts of its own InfiniBand chip design to others, said Tom Macdonald, general manager of Intel’s component division.

In addition, the company will incorporate InfiniBand technology into chipsets, the processors secondary in importance only to the CPU, meaning that InfiniBand will come standard in many Intel-based servers.

Other companies producing chips for InfiniBand include IBM, Agilent Technologies and Lucent Technologies. Intel has invested in two start-ups making InfiniBand chips – Banderacom and Mellanox – as well as in Crossroads, which is bringing InfiniBand to high-speed storage network hardware, Macdonald said at the Intel Developer Forum this week in San Jose.

The high-end technology still has a ways to go. Prototype chips are beginning to arrive, Macdonald said, but the InfiniBand specification itself isn’t expected to be final until September or October.

InfiniBand is a critical technology for Intel as it tries to produce server designs that can catch up to proprietary designs from IBM, Hewlett-Packard, Compaq Computer and especially Sun Microsystems.

First, selling InfiniBand chips and licensing intellectual property will be a revenue source. Second, in the bigger picture, InfiniBand will help the adoption of Intel servers in data centers and other demanding environments where customers are willing to pay for performance and reliability.

InfiniBand won’t just benefit Intel’s bottom line, though. More than 180 companies have joined the InfiniBand Trade Association. The association’s steering committee is run by Intel, IBM, Microsoft, Dell Computer, Sun, Compaq and HP.

The alliance has appointed Macdonald and IBM’s Tom Bradicich as co-chairmen, a reflection of the fact that the InifiniBand effort was preceded by two competing technologies, one from Intel, Sun and Dell and the other from IBM, Compaq and HP. After months of haggling, the two camps joined forces a year ago.

In the dispute, IBM, Compaq and HP had advocated a higher-speed but more expensive new technology called Future Input/Output (FIO), while Intel and its allies pressed for a conceptually similar, less expensive but less powerful system called Next-Generation I/O (NGIO). While the battle raged, servers based on Intel chips couldn’t advance at the pace of proprietary Unix servers.

The companies say the bad blood is behind them, but some of the philosophy lives on. Intel’s InfiniBand chips will initially be aimed at low-end systems using the slower, cheaper and simper “1X” version of InfiniBand, which offers a connection speed of 2.5 gbps (gigabits per second). IBM chips will start with more expensive “4X” designs, four times the speed.

Lucent’s chips target both the 1X and 4X standards, but the company said its models use two-thirds the power of competing chips. That’s important for data centers packed with thousands of chips that must be kept cool.

HP spinoff Agilent, on its own and in partnership with a Fujitsu spinoff called RedSwitch, also is working on InfiniBand chips, the company said.

InfiniBand chips fall into three categories: the “host” chips to communicate with CPUs; the chips to communicate with “target” devices such as network and storage adapters; and the high-speed switches that will connect the host and target chips.

Intel is licensing its design for the communication layer of InfiniBand to other companies, meaning that others could add it to switches or other chips to make sure their InfiniBand designs are compatible with Intel systems.

Intel also will sell InfiniBand development systems so that hardware and software companies can develop InfiniBand systems, said Jim Pappas, director of initiative marketing for Intel’s fabric components division. “This will be among the first, if not the first, platform available,” he said.

InfiniBand is expected to dramatically alter the way servers are built, Macdonald said. Today, servers typically have CPU and memory modules connected by a PCI bus to network cards and storage systems.

With InfiniBand, these functions will be split into separate boxes, continuing a trend toward specialization that the Internet has been driving.

One type of machine will house CPUs and memory. These machines will be connected via InfiniBand to other machines devoted to communicating with the network or to storage systems.

It’s a strategy appropriated from mainframe designs; one of its chief advantages is that customers can buy more of the components that they need: CPUs for computationally intensive jobs, or input-output and storage modules for talking to databases.

============================================================

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industry updates delivered to you every week!

Nvidia Showcases Work with Quantum Centers at ISC24

May 13, 2024

With quantum computing surging in Europe, Nvidia took advantage of ISC24 to showcase its efforts working with quantum development centers. Currently, Nvidia GPUs are dominant inside classical systems used for quantum sim Read more…

ISC24: Hyperion Research Predicts HPC Market Rebound after Flat 2023

May 13, 2024

First, the top line: the overall HPC market was flat in 2023 at roughly $37 billion, bogged down by supply chain issues and slowed acceptance of some larger systems (e.g. exascale), according to Hyperion Research’s ann Read more…

Top 500: Aurora Breaks into Exascale, but Can’t Get to the Frontier of HPC

May 13, 2024

The 63rd installment of the TOP500 list is available today in coordination with the kickoff of ISC 2024 in Hamburg, Germany. Once again, the Frontier system at Oak Ridge National Laboratory in Tennessee, USA, retains its Read more…

Harvard/Google Use AI to Help Produce Astonishing 3D Map of Brain Tissue

May 10, 2024

Although LLMs are getting all the notice lately, AI techniques of many varieties are being infused throughout science. For example, Harvard researchers, Google, and colleagues published a 3D map in Science this week that Read more…

ISC Preview: Focus Will Be on Top500 and HPC Diversity 

May 9, 2024

Last year's Supercomputing 2023 in November had record attendance, but the direction of high-performance computing was a hot topic on the floor. Expect more of that at the upcoming ISC High Performance 2024, which is hap Read more…

Processor Security: Taking the Wong Path

May 9, 2024

More research at UC San Diego revealed yet another side-channel attack on x86_64 processors. The research identified a new vulnerability that allows precise control of conditional branch prediction in modern processors.� Read more…

ISC24: Hyperion Research Predicts HPC Market Rebound after Flat 2023

May 13, 2024

First, the top line: the overall HPC market was flat in 2023 at roughly $37 billion, bogged down by supply chain issues and slowed acceptance of some larger sys Read more…

Top 500: Aurora Breaks into Exascale, but Can’t Get to the Frontier of HPC

May 13, 2024

The 63rd installment of the TOP500 list is available today in coordination with the kickoff of ISC 2024 in Hamburg, Germany. Once again, the Frontier system at Read more…

ISC Preview: Focus Will Be on Top500 and HPC Diversity 

May 9, 2024

Last year's Supercomputing 2023 in November had record attendance, but the direction of high-performance computing was a hot topic on the floor. Expect more of Read more…

Illinois Considers $20 Billion Quantum Manhattan Project Says Report

May 7, 2024

There are multiple reports that Illinois governor Jay Robert Pritzker is considering a $20 billion Quantum Manhattan-like project for the Chicago area. Accordin Read more…

The NASA Black Hole Plunge

May 7, 2024

We have all thought about it. No one has done it, but now, thanks to HPC, we see what it looks like. Hold on to your feet because NASA has released videos of wh Read more…

How Nvidia Could Use $700M Run.ai Acquisition for AI Consumption

May 6, 2024

Nvidia is touching $2 trillion in market cap purely on the brute force of its GPU sales, and there's room for the company to grow with software. The company hop Read more…

Hyperion To Provide a Peek at Storage, File System Usage with Global Site Survey

May 3, 2024

Curious how the market for distributed file systems, interconnects, and high-end storage is playing out in 2024? Then you might be interested in the market anal Read more…

Qubit Watch: Intel Process, IBM’s Heron, APS March Meeting, PsiQuantum Platform, QED-C on Logistics, FS Comparison

May 1, 2024

Intel has long argued that leveraging its semiconductor manufacturing prowess and use of quantum dot qubits will help Intel emerge as a leader in the race to de Read more…

Nvidia H100: Are 550,000 GPUs Enough for This Year?

August 17, 2023

The GPU Squeeze continues to place a premium on Nvidia H100 GPUs. In a recent Financial Times article, Nvidia reports that it expects to ship 550,000 of its lat Read more…

Synopsys Eats Ansys: Does HPC Get Indigestion?

February 8, 2024

Recently, it was announced that Synopsys is buying HPC tool developer Ansys. Started in Pittsburgh, Pa., in 1970 as Swanson Analysis Systems, Inc. (SASI) by John Swanson (and eventually renamed), Ansys serves the CAE (Computer Aided Engineering)/multiphysics engineering simulation market. Read more…

Intel’s Server and PC Chip Development Will Blur After 2025

January 15, 2024

Intel's dealing with much more than chip rivals breathing down its neck; it is simultaneously integrating a bevy of new technologies such as chiplets, artificia Read more…

Comparing NVIDIA A100 and NVIDIA L40S: Which GPU is Ideal for AI and Graphics-Intensive Workloads?

October 30, 2023

With long lead times for the NVIDIA H100 and A100 GPUs, many organizations are looking at the new NVIDIA L40S GPU, which it’s a new GPU optimized for AI and g Read more…

Choosing the Right GPU for LLM Inference and Training

December 11, 2023

Accelerating the training and inference processes of deep learning models is crucial for unleashing their true potential and NVIDIA GPUs have emerged as a game- Read more…

Shutterstock 1606064203

Meta’s Zuckerberg Puts Its AI Future in the Hands of 600,000 GPUs

January 25, 2024

In under two minutes, Meta's CEO, Mark Zuckerberg, laid out the company's AI plans, which included a plan to build an artificial intelligence system with the eq Read more…

AMD MI3000A

How AMD May Get Across the CUDA Moat

October 5, 2023

When discussing GenAI, the term "GPU" almost always enters the conversation and the topic often moves toward performance and access. Interestingly, the word "GPU" is assumed to mean "Nvidia" products. (As an aside, the popular Nvidia hardware used in GenAI are not technically... Read more…

Nvidia’s New Blackwell GPU Can Train AI Models with Trillions of Parameters

March 18, 2024

Nvidia's latest and fastest GPU, codenamed Blackwell, is here and will underpin the company's AI plans this year. The chip offers performance improvements from Read more…

Leading Solution Providers

Contributors

Shutterstock 1285747942

AMD’s Horsepower-packed MI300X GPU Beats Nvidia’s Upcoming H200

December 7, 2023

AMD and Nvidia are locked in an AI performance battle – much like the gaming GPU performance clash the companies have waged for decades. AMD has claimed it Read more…

Eyes on the Quantum Prize – D-Wave Says its Time is Now

January 30, 2024

Early quantum computing pioneer D-Wave again asserted – that at least for D-Wave – the commercial quantum era has begun. Speaking at its first in-person Ana Read more…

The GenAI Datacenter Squeeze Is Here

February 1, 2024

The immediate effect of the GenAI GPU Squeeze was to reduce availability, either direct purchase or cloud access, increase cost, and push demand through the roof. A secondary issue has been developing over the last several years. Even though your organization secured several racks... Read more…

The NASA Black Hole Plunge

May 7, 2024

We have all thought about it. No one has done it, but now, thanks to HPC, we see what it looks like. Hold on to your feet because NASA has released videos of wh Read more…

Intel Plans Falcon Shores 2 GPU Supercomputing Chip for 2026  

August 8, 2023

Intel is planning to onboard a new version of the Falcon Shores chip in 2026, which is code-named Falcon Shores 2. The new product was announced by CEO Pat Gel Read more…

GenAI Having Major Impact on Data Culture, Survey Says

February 21, 2024

While 2023 was the year of GenAI, the adoption rates for GenAI did not match expectations. Most organizations are continuing to invest in GenAI but are yet to Read more…

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing po Read more…

How the Chip Industry is Helping a Battery Company

May 8, 2024

Chip companies, once seen as engineering pure plays, are now at the center of geopolitical intrigue. Chip manufacturing firms, especially TSMC and Intel, have b Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire