Dell Unveils Eight-Socket HPC Box

By Michael Feldman

February 14, 2011

Dell has introduced an ultra-dense rack server built for high performance computing. Based on the latest and greatest Opteron 6100 series (“Magny-Cours”) processors, the PowerEdge C6145 puts two independent four-socket servers in a 2U chassis, offering up to 96 cores, 1 TB memory, and 10 PCIe slots.

Since AMD introduced its 6100 Opterons last May, it has been enticing server makers to build mainstream four-socket servers with the new chips. The company’s latest-generation x86 CPU was designed to eliminate the so-called “4P tax,” a premium price associated with four-socket platforms. Although a four-socket design saves on cabling, racks and floor space, until AMD came along with its 2P-4P crossover CPU, the more complex chipsets have tended to make the four-socket servers much more expensive.

AMD has apparently made its case. All the major OEMs now have four-socket offerings based on the 6100 Opterons. But until this week, none were explicitly targeted for HPC duty.

The PowerEdge C6145 is essentially the Opteron counterpart to Dell’s Intel Xeon-based PowerEdge C6100, an HPC server that houses four two-socket nodes in a 2U chassis. The C6100 is the platform that powers the 90-teraflop “Hyperion” system at Lawrence Livermore National Lab. That system is based on quad-core Nehalem EP chips, but the C6100 is also offered with the latest six-core Westmere EP CPUs.

What the C6145 brings to the table is more x86 cores and more memory capacity in the same 2U form factor. Using the 8-core 6100 Opterons, a single server gets you to 64 cores per box; and to 96 cores with the 12-core version. Memory capacity can go up to 128 GB or 256 GB per board using 4 and 8 GB DIMMs, respectively. The upcoming 16 GB DIMMs will push this to 512 GB per board or 1 TB in the dual-node configuration.

For maximum flexibility, Dell decided to cram a lot more PCIe slots into the C6145 than it had in the C6100. In the latter case, you have just two PCIe slots (an x8 mezzanine daughtercard slot and an x16 riser slot) per node, while the new C6145 sports five (3 x16 PCI slots, an x8 mezzanine slot, and an x16 host interface card, or HIC). The idea is to be able to mix and match RAID controllers, InfiniBand cards, 10GbE NICs, and HIC devices in a very dense, core-rich server.

All the extra PCIe slottage is perfect for connecting up with Dell’s PowerEdge C410x, a PCIe expansion chassis that can be outfitted with up to 16 GPGPUs. In fact, according to Tim Carroll, Dell’s global lead of research computing, their C410x customers had a lot to do with the C6145 design.

Early users of Dell’s GPU box ended up making CPU-GPU server sandwiches with the C6100 and the C410x, but according to Carroll, what they really wanted on the CPU side was higher core counts, more memory and more PCI connectivity. The core-laden 6100 Opterons and the extra space available from a four-socket design made all this possible.

It’s worth noting a bit of irony here. The C410x is currently qualified for NVIDIA Tesla GPUs, not AMD-built ones. Nonetheless, the AMD CPUs make a perfectly respectable (and vendor-agnostic) partner to the GPGPUs.

The C6145 is not just a GPGPU driver, though. Any HPC workload that benefits from more cores, memory, and bandwidth will get a boost from the Opteron 6100. Applications in life sciences, oil and gas, computational fluid dynamics, EDA, and medical imaging should be especially favored.

This could also be an attractive platform for customers looking to consolidate their HPC infrastructure with more mundane enterprise tasks. Because of the high core count and memory capacity, the C6145 is well-suited to dense enterprise virtualization and virtual desktop infrastructure. Likewise, video rendering and transcoding could offer another sweet spot for this product.

Performance-wise, the C6145 has plenty to recommend it. Configured with the top Opteron parts, the server exhibits the highest x86 floating point throughput on the planet for an 8-socket box (based on SPECfp_rate2006 result). Prior to this, the top 8-socket record holder was the Xeon-based HP DL980 G7, a higher end offering that costs about five times as much, and at 8U, takes up four times the space.

The performance advantage can be directly attributable to the 6100 Opterons, which themselves got a speed bump on Monday. In fact, the C6145 that broke the SPECfp_rate2006 record, was outfitted with the new Opteron 6180 SE CPU, a 12-core version running at 2.5 GHz. That’s 200 MHz faster than the previous high end part, the 12-core Opteron 6176 SE. AMD also announced four other newly-accelerated 6100 CPUs: the 8-core 2.2 GHz 6132 HE and 2.6 GHz 6140, and the 12-core 1.8 GHz 6166 HE and 2.3 GHz 6176.

With the new Bulldozer architecture debuting in the upcoming “Interlagos” Opterons, an even larger speed bump is on the horizon. According to AMD, the 16-core Interlagos CPU will deliver up to 50 percent greater compute throughput and memory throughput compared to the current generation 6100 Opterons. The first such chips are scheduled to begin production in Q2, with widespread availability in Q3.

Thanks to socket compatibility, the Interlagos parts are plug compatible with the 6100s, allowing system vendors to easily transition to the new CPU architecture on the same platform. According to Carroll, the PowerEdge C6145 is due for an Interlagos upgrade when they become available. At that point, Dell will be able to offer a 128-core 2U server — something that could easily have passed for a mid-sized cluster a couple of years ago.

That’s not to say Dell will stop churning out Xeon-based gear. The Intel CPUs still reign supreme when it comes to single-threaded performance. (At 3.46 GHz, the fastest Xeon X5677 is nearly a full GHz quicker than the top end Opterons.) For HPC codes that can only scale so far, the higher clocks are the way to go.

“People spend a lot less time talking about Intel versus AMD today,” notes Carroll. “They’re really more driven by dollars per flop per watt per rack. I also think it’s going to be driven by customer requirement and use case, rather than whether it’s an AMD shop or an Intel shop.”

List price for the PowerEdge C6145 starts at $18K, with availability beginning at the end of the month. The new AMD 6100 Opteron processors are available immediately, with prices ranging from $591 to $1,514 in units of a thousand.

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