Supercomputing Seeks Energy Savings

By Scott Jones

March 5, 2009

As high-performance computing (HPC) enters the petascale age, the scientific challenges facing researchers have never been greater. Nor has the might of today’s production petascale machines. The recent exponential growth in the power of modern supercomputers has gone hand-in-hand with an increased demand on resources — as machines have gotten bigger and faster, the amount of resources required for their operation has likewise increased.

As a result, HPC centers now face unprecedented power demands from the very machines they rely on to tackle today’s most daunting scientific challenges, from climate change to the modeling of biological processes. However, recent energy-saving innovations at ORNL are setting a new standard for resource-responsible HPC research. The laboratory has taken an all-angles approach, seeking energy savings from a suite of different areas.

ORNL’s leadership system, a Cray XT known as Jaguar, is now the fastest computer in the world for open science with a maximum speed of 1.6 petaflops. With this great power comes great responsibility, especially when it comes to energy consumption. “We take energy utilization very seriously,” said ORNL’s Leadership Computing Facility Project Director Buddy Bland. “The scale of this machine is just phenomenal. There are very few places in the world where this computer could have been built.”

Needless to say, feeding this animal is no small task: simulation at the petascale requires robust power and cooling networks to ensure maximum production from these machines. But now those necessary support networks, and the system itself, have been designed with unprecedented efficiency, responsibly satisfying Jaguar’s energy appetite. These advances make ORNL among the most energy-efficient locations for HPC, enabling groundbreaking research with minimal resource impact.

It all starts with the building. ORNL’s Computational Sciences Building (CSB) was among the first Leadership in Energy and Environmental Design (LEED)-certified computing facilities in the country, meaning that its design satisfies criteria used by the U.S. Green Building Council to measure the efficiency and sustainability of a building.

Take the computer room for example: it’s sealed off from the rest of the building by a vapor barrier to reduce the infiltration of humidity. The air pressure inside the computer room is slightly higher than the surrounding area so air will flow out of the computer room without the air outside flowing in.

Because ORNL is located in an area of the country with high humidity, keeping moisture out of the air is a high priority said Bland, one that the building was designed to tackle as efficiently as possible. Too much moisture in the air can lead to water condensation on equipment, while too little moisture can cause static electricity to build up — both of which can be problematic for a room filled with expensive electronics. Both removing moisture from or adding it to the air uses a lot of power, so keeping the humidity stable is a great tool for reducing energy consumption.

Another computing building on the ORNL campus adjacent to the CSB was recently certified LEED Gold, and Bland points out that the laboratory plans on an equal rating for future HPC facilities. But the innovation doesn’t stop with the building — there is plenty more under the roof.

Jaguar requires huge amounts of chilled water to keep the machine cool. To accomplish this as efficiently as possible, the laboratory uses high-efficiency chillers, which are the first step in a multifaceted, efficient cooling design.

A newly introduced Cray cooling system for Jaguar, dubbed ECOphlex, complements the chillers and the CSB’s efficiency. Using a common refrigerant and a series of heat exchangers, ECOphlex efficiently removes the heat generated by Jaguar to keep the computer room cool. The combination of air- and refrigerant-based cooling is much more efficient than traditional systems, which rely almost solely on air for temperature control. Without ECOphlex, the number of air-based units would not fit into the CSB’s computer room. This high-efficiency cooling system makes Jaguar possible.

ECOphlex also allows ORNL to reduce the amount of chilled water used to cool Jaguar by accommodating a broader inlet temperature range for the cooling water. Considering the fact that thousands of gallons of water per minute are necessary to keep Jaguar cool, a reduction in the volume of necessary chilled water means a proportionate reduction in the energy used to cool it. Simply put, warmer water can mean big energy savings for the laboratory and the taxpayer. Whereas most centers use 0.8 watts of power for cooling per every watt of power used for computing, ORNL enjoys a far more efficient ratio of 0.3 to 1, one of the lowest of all data centers measured.

Another important innovation is one that ORNL has been working on with Cray for several years. Instead of using the more common 208-volt power supply that Jaguar used in the past, the system now runs directly on 480-volt power. This seemingly “minor” change is saving the laboratory $1 million in the cost of copper used in the power cords for the cabinets. Furthermore, keeping the voltage high allows a lower current, which means lower resistance and less power turned into heat as it travels down the wires. The reduction in electrical resistance will reduce energy costs by as much as half a million dollars.

Finally, ORNL gets a little help from history. The power grid for the city of Oak Ridge was designed when the work conducted during the Manhattan Project used one-seventh of all the electricity in the country. The grid was constructed with every protection possible out of the fear that any interruption in supply would drastically set back development. The result: an extremely resilient local power grid.

Because of this grid, said Bland, Oak Ridge doesn’t need huge uninterruptable power supply (UPS) systems, which generally consume lots of electricity. However, the laboratory does have flywheel-based UPSs in case of an emergency. If there is a problem, the flywheel keeps generating power, which is a much more efficient process than conventional UPSs and therefore a greener method of supplying backup power. Because the flywheel-based UPS is mechanical as opposed to battery-operated, it also generates less waste in the long-term as battery replacement is not a concern.

While all of these steps are important, taken together they are greater than the sum of their parts. “There is no silver bullet,” said Bland. By tackling energy efficiency from multiple angles, ORNL is helping to ensure that the groundbreaking research taking place on its petascale machines is conducted as responsibly as possible, setting new standards in both HPC and energy responsibility.

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