The Coming Quad Wars

By Michael Feldman

August 17, 2007

Being in the x86 processor space has always been a double-edged sword for AMD. The market for this architecture, probably the most widely used ISA for general-purpose computing, is enormous. But so is the competition.

Intel dwarfs AMD by any measurement. In 2006, Intel’s net revenue was $35.38 billion, compared to $5.65 billion for AMD. In fact, last year Intel’s R&D spending alone — $5.87 billion — was larger the AMD’s entire revenue stream. This disparity is manifesting itself as AMD gets ready to release its much-anticipated “Barcelona” quad-core processor and Intel prepares to launch its new 45nm Penryn family.

On Tuesday, DailyTech reported that Intel is planning to release its first 45nm Penryn server chips on November 11, just nine weeks after the expected September 10 Barcelona release (and just in time for the 2007 Supercomputing Conference and Expo). Apparently the Penryn information was inadvertently placed on an unprotected Intel website. The 45nm shrink represents the next “tick” in Intel’s “tick-tock” processor development strategy.

With AMD just beginning its 65nm Opteron deliveries at almost the same time that Intel is introducing its 45nm Xeon chips, it looks like AMD will be nearly a full process technology cycle behind its rival. While transistor size and technology isn’t everything, it does offer the leader some fundamental advantages. For example, getting to 65nm ahead of AMD helped Intel deliver quad-core processors almost a year in advance of its rival. AMD is quick to remind us that the Intel designs are not “true” quad-core processors, since they rely on plugging two dual-core chips into a single socket. But that’s of little consequence to customers. The dual-dual chips achieve respectable performance numbers. The new Xeon L5335 Intel just announced this week is a 2.0GHz quad that consumes only 50 watts — a mere 12.5 watts per core.

While AMD is still ahead of the game in system design, using HyperTransport and an integrated memory controller to achieve better multicore integration and energy efficiency, the lack of a quad-core offering over the last ten months created an opportunity for Intel to retake market share, especially in the server space. With the quad-core Barcelona, AMD has a chance to recover some lost momentum. To its detriment, the company has over-hyped its new product and its vision of dealing a death blow to the Intel quads will not be realized. AMD scaled back the initial chip to a modest 2.0GHz, which is significantly less that the original target of 2.6GHz.

Because of the slower clock, AMD has backed away from claims of integer performance superiority for its initial Barcelona offering. However, the company still expects to beat the current raft of Xeons in floating point performance. It’s all but impossible to get an apples-to-apples comparison of processors these days, given the variation of CPU caches, clocks and power envelopes. But assuming the first Opteron quad-core will be a 2.0GHz processor at 95 watts, a comparable chip in the Intel stable might be the 2.6GHz Xeon X5355 quad-core, but which runs a hotter at 120 watts (and that doesn’t include the off-chip memory controller). AMD is claiming a 2.0GHz Barcelona should yield a peak SPECfp_rate2006 result of 69.5; Intel reports 58.9 for the Xeon X5355. So if the first quad-core Opteron out of the chute can put up these kinds of numbers against a comparable Xeon and use less power, that bodes well for the new Opteron line, especially in the floating-point-loving high performance technical computing market.

But if Intel does release Penryn-based Xeons a couple of months after the Barcelona launch, AMD’s performance edge will be in jeopardy. Because the Penryn processors will use the advanced high-K dielectric and metal gate transistor design for its 45nm manufacturing technology, these chips will be able to achieve significantly better performance within a given power envelope compared to their 65nm counterparts. I’m guessing that the Penryn launch will nullify any floating point performance advantage AMD will achieve with their first quad-core Opterons and widen Intel’s integer performance advantage. AMD, of course, will not be sitting still. They’re expected to come out with both higher performing and lower power versions of their quad offering following the initial September introduction. So by the time November rolls around, we could have a real horse race.

But being an entire process technology iteration behind their rival will be a heavy burden for AMD. Intel will have a lot of latitude in targeting performance or energy consumption with the new transistor technology. If we can believe the Penryn information leaked by Intel, the top of the line X5460 quad-core will clock in at 3.16GHz and dissipate 120 watts; the low-end quad-core L5410 will run at 2.33GHz and use just 50 watts. Until AMD can move to 45nm, which it plans to do in 2008, it will have to be content to find design tradeoffs where it can tweak the clock speed or exploit energy savings on the 65nm chips.

The only good news for AMD is that the Penryn chips will still rely on the antiquated Front Side Bus (FSB) and off-chip memory controllers. However, this will not be the case for Intel’s next-generation Nehalem microarchitecture due in 2008 — the next “tock” in the tick-tock strategy. Nehalem is expected to jettison the FSB in favor of a more HyperTransport-like system interconnect and use an integrated memory controller. There’s even talk of including a GPU on some Nehalem processors. The fact that Intel is following in its smaller rival’s footsteps is probably of little consolation to AMD at this point.

For AMD to compete effectively in the x86 server market, it has to be on par with Intel’s semiconductor technology. That doesn’t mean AMD has to embrace the same technologies or have the same manufacturing structure as Intel, but if Opteron transistors are always going to be bigger and leakier than Xeon transistors, that seems unsustainable in the long run.

Since AMD relies heavily on IBM for its process technology R&D, maybe it’s time to invest more heavily in that partnership — or develop a larger alliance of companies that it can use for both semiconductor research and manufacturing. Big Blue itself would not appear to be as motivated as AMD is in outgunning Intel chips, since IBM’s own Power and PowerPC processors aren’t in direct competition with the Xeons. (Itaniums, of course, are another matter.) Indeed, IBM manufactures both Xeon- and Opteron-based servers. That means AMD itself needs to find a way to drive the process. But time is not on their side. Tick-tock.

—–

As always, comments about HPCwire are welcomed and encouraged. Write to me, Michael Feldman, at [email protected].

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