Cray Unveils New Cooling Technology For the Petascale Era

By Michael Feldman

August 28, 2008

Cray, known for its power and packaging prowess since 1976, when Seymour Cray bent the Cray-1 into a “C” shape, is unveiling a petascale-era cooling technology it says is more than 10 times as efficient as same-size water coils. Cray CTO Steve Scott discusses this innovation and the company that was green before green was cool.

HPCwire: What is Cray’s new cooling technology?

Scott: We call it ECOphlex technology. The “phlex” part refers to multiple things. First, the cabinet infrastructure can use either Cray’s high-efficiency vertical air cooling or our new phase change cooling technology that converts an inert refrigerant, R134a, from a liquid to a gas. The other flexibility is that the liquid-cooled systems can use various chilled or unchilled datacenter water temperatures to pull heat from the R134a subsystem and to adapt to changing datacenter conditions. The phase change coil is more than 10 times as efficient at removing heat from the compute cabinets as a water coil of similar size, so the in-cabinet cooling system is much smaller and lighter than it would be with water coils. Water is only used in external heat-exchange units.

The ECOphlex technology is the first of the Cray “Baker” technologies we’re introducing. We’ll start using it when we ship the Cray XT5 petascale system to Oak Ridge later this year. After that, all Cray XT5 systems will ship with ECOphlex capability in the new high-efficiency cabinet.

HPCwire: With system sizes and densities increasing, liquid cooling is making headlines as if it were something new, but Cray’s been at this a long time.

Scott: We’ve implemented six different types of liquid cooling since 1976, along with multiple air-cooled implementations. The Cray-1 used Freon with copper cold plates. Then we moved to fluorinert immersion, captive fluorinert cold plates, water cap cooling on the MTA-2, spray evaporative cooling on the X1, a water-cooled radiator on the X2, and the phase change liquid cooling on the XT5 series. We’ve gained a lot of experience with what forms of liquid cooling work best under various constraints.

HPCwire: How does your phase change cooling compare with the typical chilled water scenario datacenters use today?

Scott: Typically today, the computers put the heat into the air, and then the CRAC [Computer Room Air Conditioner] units around the room periphery have to remove the heat from the air and put it into the chilled water. This method is very inefficient. For a petascale system the area taken up by the CRAC units could exceed the computer footprint, and this would also waste a lot of power. Cray’s new cooling scheme puts the heat into a refrigerant stream inside the rack, and then sends it to an Extreme Density Pumping unit that efficiently transfers the heat to the building chilled water. So, you still use chilled water, but it’s much less extensive and it doesn’t intrude on the computational components of the system.

ECOphlex technology is designed to be “room air neutral” within plus or minus 10%. We’ve demonstrated the ability to remove up to 100 kilowatts from a single cabinet. A typical installation would be configured with just a few CRAC units for humidity control or to deal with some leaking from other devices. The ECOphlex technology requires only a small temperature delta in the water supply, so in cooler climates or where datacenters can run at warmer ambient room temperatures, there is the potential to completely eliminate the need for expensive water chillers.

Another advantage is that since ECOphlex uses an inert coolant, you don’t have to worry about water leakage or condensation that could damage electronic components. As you know, this can be a severe problem with intrusive water-cooling technologies that bring the water-cooling close to heat-generating computer components.

HPCwire: HPC vendors have been working to advance system densities and power and cooling abilities since the start of the supercomputing era. What’s changed lately?

Scott: Supercomputing began in an era of cheap electicity. When Seymour Cray bent the Cray-1 into a “C” shape and cooled it with Freon and copper cold plates, the goal was to improve performance. Today, when Cray and other HPC vendors improve densities and power and cooling technologies, it’s not just to boost performance. It’s also for environmental reasons. We’re trying to help customers save on very expensive energy costs and facility space.

HPCwire: What about other approaches to reducing energy use and heat generation, such as using lower-power processors or accelerators?
 
Scott: There are two basic approaches. In the first, you drop the voltage and lower the frequency of individual processors, then compensate by using more processors in a system. Multi-core processors embody this approach to a moderate extent, and some special purpose designs have taken it even farther. The primary concern here is that this approach exacerbates the scaling problem. The memory wall gets worse, there’s more memory contention, codes have to be more parallel, the communication-to-computation ratio gets worse, and you have to depend more on locality. This approach works well for highly localized, partitionable applications. The more you push this concept, however, the more potential power savings you have for certain codes, but the more special-purpose the machine becomes.

Another alternative is to design processors that have much lower control overhead and use more of their silicon area for performing computations. Streaming processors, vector processors and FPGAs are examples of this approach, which can result in much faster single processors for the right types of codes, and thus ease the requirement for greater scaling. This technique can be used to a lesser extent in traditional scalar microprocessors. SSE instructions, for example, are essentially vector instructions that can increase peak performance without a corresponding increase in control complexity. On top of all this, you can also implement adaptive power-management mechanisms to reduce power consumption by idling or voltage scaling selected blocks of logic in the processor. Microprocessor vendors have a big motive to reduce power consumption because it affects their whole market, not just the relatively small HPC segment.

HPCwire: So which techniques do you think hold the most promise?

Scott: I don’t think there’s one right answer. Ultimately, the important thing is matching the capabilities of the machine with the needs of the applications. The variety of applications calls for a variety of solutions, each optimized for the right system balance. This will lead to more performance efficiency and power efficiency. What you don’t want to do is compromise application performance. In the end, it’s watts per sustained performance that matters, not watts per peak performance.

HPCwire: Are there any other important aspects of Green HPC?

Scott: Yes. Another really important dimension is equipment disposal, though this one doesn’t get as much attention as power and cooling. In many cases, buying a new supercomputer today requires a forklift upgrade of the cabinets. This is true of some of the most efficient systems on the Green500 list, which just looks at kilowatts per flop. Recycling cabinets every two to three years, or trucking them to a landfill, isn’t very environmentally friendly and can cost extra money for the customer on both the initial system purchase and the disposal. Cray has been using multi-generational cabinets with our XT series, and many of our customers have already gone through two or three processor upgrades in the same cabinets. Our new high-efficiency cabinet continues this practice.

Another important factor is power efficiency inside the cabinet. Since the Cray XT3, we’ve been using a single axial turbofan in our cabinets. It’s a lot more efficient than a large collection of less-powerful fans and the maintenance interval is seven-and-a-half years, versus a few months with the small fans.

We’re also using AC/DC power rectification with a 90%-plus efficiency rating. There’s much less power loss with these power supplies, and they can support higher-power processors.

HPCwire: Any parting thoughts?

Scott: Just that as the HPC community begins to enter the petascale computing era, the challenge of cooling large-scale systems and paying for the energy costs is escalating very quickly. Cray and every other vendor serving this market will need to push “green” innovation hard to stay ahead of this curve for customers. We’re fortunate at Cray to have a long history of power, cooling and packaging innovation, and we’re committed to do what it takes, in cooling and other areas, to enable our customers to get to sustained petascale computing capabilites.

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!

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…

Nvidia Appoints Andy Grant as EMEA Director of Supercomputing, Higher Education, and AI

March 22, 2024

Nvidia recently appointed Andy Grant as Director, Supercomputing, Higher Education, and AI for Europe, the Middle East, and Africa (EMEA). With over 25 years of high-performance computing (HPC) experience, Grant brings a 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…

Houston We Have a Solution: Addressing the HPC and Tech Talent Gap

March 15, 2024

Generations of Houstonian teachers, counselors, and parents have either worked in the aerospace industry or know people who do - the prospect of entering the fi 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