Chips Ahoy: Vendors Show Off Their Latest Silicon

By Michael Feldman

February 9, 2010

Chipmakers converged on San Francisco this week to talk up their newest semiconductor products at the International Solid State Circuits Conference (ISSCC). Of particular interest to the HPC crowd are Intel’s Westmere EP and “Tukwila” Itanium 9300, and IBM’s POWER7.

In truth, the new quad-core Tukwila is not likely to have much of an impact on HPC. SGI is the only real hope that this seventh-generation Itanium will see any supercomputing action. Under the new Rackable leadership, SGI has left a lot to the imagination as far as possible Itanium-equipped Altix systems. SGI previewed its Nehalem EX-based Altix UV servers last November, and implied there would be Itanium-based versions of UV at some point, but has yet to talk about any products.

It might be relatively straightforward for SGI to build a Tukwila-based UV. The Itanium 9300 processors share platform components with Nehalem EX, including the QuickPath Interconnect (QPI), the Scalable Memory Interconnect, the 7500 Memory Buffer, and the I/O hub. Because of this commonality, Intel says manufacturers could use a common node controller for both Nehalem EX and Itanium 9300 systems. Given that SGI has already built a UV hub node controller for its shared memory systems, the company may have an easy path to an Itanium UV product.

But in general, the new Itanium is being targeted for mission-critical systems in the enterprise. These are typically high-end servers that can’t tolerate any downtime, and are especially valued for high-volume transactional applications in industries like energy, health care, telecom and manufacturing. According to Intel, Itanium’s penetration into this market is growing, reaching $5 billion in 2008 (estimated to be $4 billion in 2009 due to the recession). The chipmaker also points to a growing roster of OEMs that will be offering Itanium 9300-based machines, including Bull, HP, NEC, Hitachi, and new Itanium converts, Supermicro and China-based Inspur.

By contrast, the Westmere EP is guaranteed to see plenty of HPC action. The new Xeon chip is the 32 nm shrink of the highly popular quad-core Nehalem EP for dual-socket servers. Intel’s x86 franchise is represented by nearly 400 of the top 500 HPC systems in the world, a proportion that is likely even higher in the overall HPC server space. Intel hasn’t locked down a date when the new Xeons will start shipping, although the plan is to get them on the street in the first half of 2010.

The new features of Westmere can be summed up thusly: six cores and 12 MB of cache. That represents a 50 percent increase compared to Nehalem EP. The smaller transistor geometries mean Intel engineers were able to cram over a billion transistors on the die, which is apparently enough silicon real estate to add the two additional cores and 4 MB more cache. A quad-core variant of the Westmere EP will also be available at some point.

Even with the additional cores and cache, there was some spare silicon left over to add support for special AES (Advanced Encryption Standard) instructions, which, as its name implies, is aimed at speeding up encryption/decryption software. The engineers also came up with some additional power gating smarts to Westmere, allowing the processor to shut down processor components other than the actual processor cores (like the L3 cache, QPI interfaces, and memory controller), although it’s not clear if this feature will be available in the Xeon server parts.

Since Intel did its big architectural reset last year with the Nehalem redesign, all the goodies from that generation — integrated memory controller, QPI interface, “Hyper-Threading,” etc. — will be carried over to the Westmere processor. That should guarantee socket compatibility with the chipsets and DDR3 memory used in the Nehalem EP machines. Whether or not this means HPC users will be swapping out Nehalem EP parts with their Westmere counterparts remains to be seen.

Finally, IBM officially launched its much-anticipated POWER7 processor this week. The new chips are aimed at high-end enterprise and supercomputing servers, and also support large-scale transaction processing and analytics workloads across all application domains. In conjunction with the chip launch, four POWER7-equipped server systems were also announced: the Power 780, 770, 755, and 750. “These are the most flexible systems ever made by any company in the world,” boasted Ross Mauri, general manager of IBM Power Systems.

Hyperbole aside, of the three chips mentioned in this article, the POWER7 is the definite performance leader. In a clear departure from the POWER6 design, which delivered high clock speeds (up to 5 GHz), dual-core processors and off-chip L3 cache, the POWER7 retreats a bit on the clock speed (3 to 4 GHz), but comes with up to 8 cores and 32 MB of on-chip L3. Compare this to Tukwila at 4 cores and 24 MB of L3, and Westmere EP at 6 cores and 12 MB. Note that both the Intel chips execute up to two threads per core simultaneously, while the POWER7 can go up to four threads. The comparison with Westmere is especially interesting since IBM managed to get two more cores, two more threads per core, and 20 more megabytes of L3 cache using roughly the same number of transistors: 1.2 billion for POWER7 versus 1.17 billion for Westmere EP.

So how did Big Blue manage to make the most of its die real estate? The biggest contributor was IBM’s decision to go with embedded DRAM (eDRAM) for the on-chip L3 cache. Compared to traditional SRAM-based L3, which uses six transistors per bit, eDRAM uses just one transistor plus one capacitor. According to IBM, if they relied on SRAM technology, the equivalent chip would have consumed around 2 billion transistors and used significantly more power.

Given the x86 juggernaut in high performance computing, it’s not clear how much of the market POWER7 will grab. It was interesting that IBM’s press release included a podcast with Cindy Farach-Carson, associate vice provost for Research at Rice University and a professor of biochemistry and cell biology, who was introduced as an early user of POWER7 technology. Her work involves analyzing cancer genomic data to find the micro-RNA sequence responsible for turning slow growing cancers into more invasive and deadly variants.

The Power 755 server is the POWER7 product IBM has built for the HPC market. A 755 box contains four POWER7 processors, and since each core can execute up to four threads, a single node has the capability to run 128 threads simultaneously. Presumably this is the server (or a version thereof) that will go into the multi-petaflop Blue Waters supercomputer destined for the University of Illinois at Urbana-Champaign/NCSA sometime in 2011. In the interim, IBM is hoping other HPC users latch on to POWER7. If not, IBM will be happy to sell you a Westmere EP cluster.

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!

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion XL — were added to the benchmark suite as MLPerf continues 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 power it brings to artificial intelligence.  Nvidia's DGX Read more…

Call for Participation in Workshop on Potential NSF CISE Quantum Initiative

March 26, 2024

Editor’s Note: Next month there will be a workshop to discuss what a quantum initiative led by NSF’s Computer, Information Science and Engineering (CISE) directorate could entail. The details are posted below in a Ca Read more…

Waseda U. Researchers Reports New Quantum Algorithm for Speeding Optimization

March 25, 2024

Optimization problems cover a wide range of applications and are often cited as good candidates for quantum computing. However, the execution time for constrained combinatorial optimization applications on quantum device Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at the network layer threatens to make bigger and brawnier pro Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HBM3E memory as well as the the ability to train 1 trillion pa Read more…

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion 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…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HB Read more…

Nvidia Looks to Accelerate GenAI Adoption with NIM

March 19, 2024

Today at the GPU Technology Conference, Nvidia launched a new offering aimed at helping customers quickly deploy their generative AI applications in a secure, s Read more…

The Generative AI Future Is Now, Nvidia’s Huang Says

March 19, 2024

We are in the early days of a transformative shift in how business gets done thanks to the advent of generative AI, according to Nvidia CEO and cofounder Jensen 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…

Nvidia Showcases Quantum Cloud, Expanding Quantum Portfolio at GTC24

March 18, 2024

Nvidia’s barrage of quantum news at GTC24 this week includes new products, signature collaborations, and a new Nvidia Quantum Cloud for quantum developers. Wh Read more…

Alibaba Shuts Down its Quantum Computing Effort

November 30, 2023

In case you missed it, China’s e-commerce giant Alibaba has shut down its quantum computing research effort. It’s not entirely clear what drove the change. 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…

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…

DoD Takes a Long View of Quantum Computing

December 19, 2023

Given the large sums tied to expensive weapon systems – think $100-million-plus per F-35 fighter – it’s easy to forget the U.S. Department of Defense is a 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…

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…

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…

Baidu Exits Quantum, Closely Following Alibaba’s Earlier Move

January 5, 2024

Reuters reported this week that Baidu, China’s giant e-commerce and services provider, is exiting the quantum computing development arena. Reuters reported � Read more…

Leading Solution Providers

Contributors

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…

Shutterstock 1179408610

Google Addresses the Mysteries of Its Hypercomputer 

December 28, 2023

When Google launched its Hypercomputer earlier this month (December 2023), the first reaction was, "Say what?" It turns out that the Hypercomputer is Google's t 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…

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…

Google Introduces ‘Hypercomputer’ to Its AI Infrastructure

December 11, 2023

Google ran out of monikers to describe its new AI system released on December 7. Supercomputer perhaps wasn't an apt description, so it settled on Hypercomputer Read more…

China Is All In on a RISC-V Future

January 8, 2024

The state of RISC-V in China was discussed in a recent report released by the Jamestown Foundation, a Washington, D.C.-based think tank. The report, entitled "E Read more…

Intel Won’t Have a Xeon Max Chip with New Emerald Rapids CPU

December 14, 2023

As expected, Intel officially announced its 5th generation Xeon server chips codenamed Emerald Rapids at an event in New York City, where the focus was really o Read more…

IBM Quantum Summit: Two New QPUs, Upgraded Qiskit, 10-year Roadmap and More

December 4, 2023

IBM kicks off its annual Quantum Summit today and will announce a broad range of advances including its much-anticipated 1121-qubit Condor QPU, a smaller 133-qu Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire