NSF Official On New Supers, Data-Intensive Future

By Nicole Hemsoth

March 28, 2013

It has been a noteworthy week in the world of scientific and technical computing as two long-awaited supercomputers have been formally revved up for big research action.

The Dell-Intel scientific workhorse, Stampede, at TACC was ushered into the large-scale distributed research fold yesterday. And at the moment of this writing, the rather storied IBM and then Cray-backed Blue Waters system at NCSA is gearing up for its formal intro.

At the heart of both of these systems is some serious monetary backing from the National Science Foundation (NSF), which has committed several million to seeing both supers into the world—no matter how entangled the path. The organization funded the large majority of both projects in the name of furthering some critical human-centered scientific projects related to the environment, gemonics, disaster preparedness and epidemiology.

We chatted earlier this week with Alan Blatecky who directs the NSF’s Division of Advanced Cyberinfrastrcture about where these supers fit into the overarching mission of the NSF–and what the future looks like as applications require systems that are as “big data” ready as they are computationally robust.

Blatecky reiterated that from an NSF standpoint, these are two major investments in HPC, but they aren’t necessarily related in terms of anticipated use or application types. As he told us, the two systems are designed for quite different purposes.

One the one hand, the massive Stampede will cater to a large number of users with an emphasis on boosting the breadth of applications—not to mention extending what those extended apps are able to crunch. Blue Waters, on the other hand, will focus on a much smaller number of users, perhaps as many as a dozen, who have very deep, specific research applications.

While grappling with multiple users across a distributed system like Stampede and its XSEDE base is never simple, there are far more pressing challenges. In addition to pointing to extensive application retooling that needs to happen, especially on Blue Waters, there was one phrase we heard several times–“big data”.

The ability to take advantage of the large number of cores on a machine like Blue Waters is one of the biggest challenges user will face, says Blatecky, who points to how his organization is providing support on the programming and computer science front to aid domain specialist scientists. He said that going forward, the systems that will shine for the “big science” endeavors of the NSF will be those that can strike a balance between being data-intensive systems while retaining the computational power of massive numbers of cores, some of which are being pushed by accelerators and co-processors.

As Blatecky detailed, “Our point of view at the NSF is focused on the broader base of scientific users. We’re interested in the data-intensive computational requirements, which is part of what’s unique about Blue Waters. It has that needed balance between power, memory and storage to address both the data-intensive and computationally-intensive applications.”

When asked about the supercomputing goals the NSF wants to support over the next five years, Blatecky said that the real mission is to support a broader group of scientific users, especially those working in hot applications like genomics, materials science and environmental research areas. Most of their plans revolve around socially-oriented missions, including studies to predict earthquakes, flood outcomes, disaster response situations, and medically-driven research on the HIV and epidemic modeling fronts.

We also talked briefly about how HPC as we know it–and the NSF funds it–could change over the next five years. “I don’t know what it will be,” he noted, but he has no doubt that the performance-driven architectures might not be enough to keep up with the very real data explosion across real science applications unless they strike that memory/storage/power balance that Blue Waters has.

While not all HPC application are necessarily hugely data-intensive, a look down the list of applications reveals some of the highest data volume-driven research areas in science, particularly around medical and earth sciences projects. TACC, for instance, will now be the center of some cutting-edge earthquake, environmental and ecological research as scientists from around the world bring their best and brightest ideas –not to mention an unprecedented level of data–to the common table of the shared resource.

As TACC Director Jay Boisseau stated upon the formal announcement of Stampede yesterday, the system has been “designed to support a large, diverse research community. We are as excited about Stampede’s comprehensive capabilities and its high usability as we are of its tremendous performance.” On that note, 90% of TACC’s new powerhouse will be dedicated to the XSEDE program, which is a unified virtualized system that lets global scientists tap into powerful systems, new data wells and computational tools through one hub.

TACC will tap into the remaining horsepower for larger goals within its own center and in the University of Texas research community. And there is certainly some power to the system. As TACC described cleanly in their own statement on the specs, the Dell and Intel system boasts the following points of pride:

Stampede system components are connected via a fat-tree, FDR InfiniBand interconnect. One hundred and sixty compute racks house compute nodes with dual, eight-core sockets, and feature the new Intel Xeon Phi coprocessors. Additional racks house login, I/O, big-memory, and general hardware management nodes. Each compute node is provisioned with local storage. A high-speed Lustre file system is backed by 76 I/O servers. Stampede also contains 16 large memory nodes, each with 1 TB of RAM and 32 cores, and 128 standard compute nodes, each with an NVIDIA Kepler K20 GPU, giving users access to large shared-memory computing and remote visualization capabilities, respectively. Users will interact with the system via multiple dedicated login servers, and a suite of high-speed data servers. The cluster resource manager for job submission and scheduling will be SLURM (Simple Linux Utility for Resource Management).

Unlike Stampede, which is expected to make a top 5 showing on the Top 500m Blue Waters will not be benchmarking for reasons NCSA’s Bill Kramer explained to us in detail right around SC12. Of course, not that it needs to convince us that it will be a scientific powerhouse..

The Blue Waters saga began back in 2007 when the NSF funded the super to the tune of $208 million. At the time, IBM was at the heart of the project but refunded their payments for Blue Waters system  after looking at the cost versus return equation. Cray was later selected to take over the project with a $188 million contract that would lead the super into completion.

In the year since the video below was filmed, work on the system was completed and Blue Waters was installed at NCSA. The 11.6 petaflops (peak) supercomputer contains 237 XE cabinets, each with 24 blade assemblies, and 32 cabinets of the Cray XK6 supercomputer with NVIDIA Tesla GPU computing capability.

Currently available in “friendly-user” mode for NCSA-approved teams, Blue Waters provides sustained performance of 1 petaflop or more on a range of real-world science and engineering applications.

“Blue Waters is an example of a high-risk, high-reward research infrastructure project that will enable NSF to achieve its mission of funding basic research at the frontiers of science,” said NSF Acting Director Cora Marrett.  ”Its impact on science and engineering discoveries and innovation, as well as on national priorities, such as health, safety and well-being, will be extraordinary.”

What follows are a few examples of the exciting and promising research on Blue Waters (following provided by the National Science Foundation).

Modeling HIV

Blue Waters is enabling Klaus Schulten and his team at UIUC to describe the HIV genome and its behavior in minute detail, through computations that require the simulations of more than 60 million atoms.  They just published a paper in PLOS Pathogens touting an early discovery–not (yet) the structure of the HIV virus, but that of a smaller virus, which could only be achieved through a 10 million atom, molecular dynamics simulation, inconceivable before Blue Waters. The team is using Blue Waters to investigate complex and fundamental molecular dynamics problems requiring atomic level simulations that are 10 to 100 times larger than those modeled to date, providing unprecedented insights.

Global Climate Change

Also featured at the dedication event, Cristiana Stan and James Kinter of George Mason University are using Blue Waters to engage in topical research on the role of clouds in modeling the global climate system during present conditions and in future climate change scenarios.

Earthquake Prediction

A team at the Southern California Earthquake Center, led by Thomas Jordan, is carrying out large-scale, high-resolution earthquake simulations that incorporate the entire Los Angeles basin, including all natural and human-built infrastructure, requiring orders of magnitude more computing power than studies done to date. Their work will provide better seismic hazard assessments and inform safer building codes:  Preparing for the Big One.

Flood Assessment, Drought Monitoring, and Resource Management

Engineering Professor Patrick Reed and his team from Penn State, Princeton and the Aerospace Corporation, are using Blue Waters to transform understanding and optimization of space-based Earth science satellite constellation designs.  “Blue Waters has fundamentally changed the scale and scope of the questions we can explore,” he said.  “Our hope is that the answers we discover will enhance flood assessment, drought monitoring, and the management of water resources in large river basins world-wide.”

Fundamental Properties of Nature

Robert Sugar, professor of physics at the University of California, Santa Barbara is using Blue Waters to more fully understand the fundamental laws of nature and to glean knowledge of the early development of the universe. ”Blue Waters packs a one-two punch,” said Sugar, “Blue Waters enables us to perform the most detailed and realistic simulations of sub-atomic particles and their interactions to date. Studies such as these are a global endeavor, and the large data sets produced on Blue Waters will be shared with researchers worldwide for further discoveries.”

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