Examining the International Computational Ecosystem

By John E. West

May 7, 2009

At the end of April, the World Technology Evaluation Center (WTEC) released its latest report, the International Assessment of Research and Development in Simulation-Based Engineering and Science (SBE&S). The report was commissioned in 2007, and funded by funded by the National Science Foundation, the Department of Defense, the National Aeronautics and Space Administration, the National Institutes of Health, the National Institute of Standards and Technology, and the Department of Energy. The final product is 400 pages long, but don’t worry: all but about 130 pages of that is appendix. The full report [PDF] is available for download from the WTEC site.

The report is unreserved in its endorsement of the power of computing to make the world a better place: “No field of science or engineering exists that has not been advanced by, and in some cases transformed by, computer simulation.” From this perspective the authors and study sponsors set upon a research effort to figure out how key elements of the computational infrastructure are shaping up, what needs to be done to kick start lagging elements, and how the US compares to the rest of the world.

This last bit rings throughout the report, and frankly I found it distracting. I’ll say up front that I understand that the emphasis on US competitiveness is the practical result of the study sponsors needing to influence lawmakers to increase their budgets. And this will probably work, and the result will be a Good Thing. But the report’s emphasis leans more toward the position that lack of US leadership in every area of computational science is undesirable on its face. This view rings hollow in its extreme. I would much rather have seen the report, and the country for that matter, focus on doing the right thing for the development of a robust computational ecosystem, secure in the knowledge that competitive advantage flows naturally and sustainably from a long-term commitment to technical excellence coupled with a strong strategic vision. You don’t become world class by setting out to be world class solely for the sake of being world class. But enough of that.

The report looks at the status and trends in research in simulation-based engineering and science (SBE&S) throughout the US, Europe, and Asia (predominantly Japan and China).

A panel of experts reviewed and assessed the state of the art as well as levels of activity overseas in the broad thematic areas of SBE&S in life sciences and medicine, materials, in energy and sustainability; and in the crosscutting issues of next generation hardware and algorithms; software development; engineering simulations; validation, verification, and uncertainty quantification; multiscale modeling and simulation; and education.

The panel held a US workshop and then visited 59 sites in Europe and Asia, studied the literature, and read a whole pile of research reports to get a handle on the various types of activities being pursued. The study highlights findings in each of the three thematic areas and then in the crosscutting areas. I’ll focus on the crosscutting areas, since these are the meat and potatoes of high end technical computing.

Generally speaking, the panel noted that the cost of entry is very low for SBE&S, and that because “anyone can do it,” the spoils will go to those who can do it more effectively before anyone else. They also join the legion of blue ribbon reports to note that computational education everywhere is, er, let’s just say inadequate:

Inadequate education and training of the next generation of computational scientists threatens global as well as U.S. growth of SBE&S…. Nearly universally, the panel found concern that students use codes primarily as black boxes, with only a very small fraction of students learning proper algorithm and software development, in particular with an eye towards open-source or community code development.

They also reiterate the assertion that SBE&S funding in the US is too low with respect to massive commitments in Europe and Asia that are already beginning to bear fruit. For example:

In Germany, specific and focused investments in SBE&S are patterned along the recommendations in the 2006 NSF blue ribbon panel report on SBES (Oden 2006) as part of the 20+% year-on-year increase in funding for research. As a consequence of this new funding, Germany already exhibits many of the innovative organizational and collaborative structures deemed to be the most promising for advancing SBE&S in the context of energy, medicine, and materials research. The panel observed extensive restructuring of universities to enable more interdisciplinary work and strong university-industry partnerships.

Throughout the report the panel criticizes the US short-term funding emphasis, the lack of strategic commitment, and our cultural “go it alone” attitude as determinants of perpetual weakness in our computational infrastructure.

For example, the panel found that community code development projects are much stronger within the European Union than the United States, with national strategies and long-term support. Many times the panel was told that the United States is an “unreliable partner” in these efforts due to our inability to commit for longer than typically three years at a time. Both this perception and the reality means that the United States has little influence over the direction of these community codes and at the same time is not developing large codes of its own.

And there is this gem, aimed at materials science efforts in the US, but applicable to HPC in general:

Many large codes, both open source and non-open source, require collaboration among large groups of domain scientists, applied mathematicians, and computational scientists. However, there is much greater collaboration among groups in materials code development in Europe compared to the United States. There appear to be several reasons for this:

  • The U.S. tenure process and academic rewards systems suppress collaboration.
  • Funding, promotion, and awards favor high-impact science (publications in Nature and Science, for example), while the development of simulation tools is not considered to be high-impact science. Yet, these tools (which can take many years to develop) are often the key factor in enabling the high-impact science.

But what about HPC in specific, you ask? The panel re-emphasizes the observations of many other panel reports in pointing out the value of coordinated, national-scale investments in computational infrastructure:

The many orders-of-magnitude in speedup required to make significant progress in many disciplines will come from a combination of synergistic advances in hardware, algorithms, and software, and thus investment and progress in one will not pay off without concomitant investments in the other two.

The authors observe that there are healthy investments throughout Europe, Asia, and the US aimed at developing and sustaining high end computing hardware, but stops short of endorsing them at currently-funded levels. Interestingly, the report also acknowledges that simulation-based engineering and science software thrives on a continuum of hardware, from supercomputers to desktops, and that tools, frameworks and computing platforms need to be provisioned that recognize and support this need. This is not often discussed, but rings true in my own experiences supporting a large, diverse user community.

Immature tools and the higgeldy-piggeldy nature of the HPC workflow in general are called out as presenting substantial hurdles to more effective use of SBE&S.

Software and data interoperability, visualization, and algorithms that outlast hardware obstruct more effective use of engineering simulation…. In most engineering applications, algorithms, software and data/visualization are primary bottlenecks. Computational resources (flops and bytes) were not limiting factors at most sites. Lifecycle of algorithms is in the 10-20 years range, whereas hardware lifecycle is in the 2-3 years range. Visualization of simulation outputs remains a challenge and HPC and high-bandwidth networks have exacerbated the problem. (Emphasis added)

In particular, investments in data workflow and large-scale visualization are found by the report to be unevenly distributed, with the particle physics and biological sciences communities leading the way, and chemical and material science communities bringing up the rear. In part, the leading communities are ahead because they haven’t been allowed to simply limp along with the way they’ve always done it: regulatory requirements (in biological science) and the sheer volume of data have forced investment and innovation. The other communities are still largely trading off the immediate waste of researcher time against the longer-term investment in a supporting data infrastructure that would ultimately accelerate the pace of innovation and discovery. In other words they are buying more flops now and sacrificing more discovery later. But, the report finds:

Industry is significantly ahead of academia with respect to data management infrastructure, supply chain, and workflow.

The report also finds that visualization and data analysis are completely essential to solving our society’s most important problems, directly addressing the disturbing trend in some large-scale national HPC programs to de-emphasize funding for visualization, and to paint its contribution as “pretty pictures”:

Big data, visualization and dynamic data-driven simulations are crucial technology elements in numerous “grand challenges,” including the production of transportation fuels from the last remaining giant oil fields.

Finally, the panel’s report highlights a huge hole in the global computational ecosystem: the lack of proper emphasis on verification and validation and uncertainty quantification.

A report on European computational science (ESF 2007) concludes that “without validation, computational data are not credible, and hence, are useless.”…The data and other information the WTEC panel collected in its study suggests that there are a lot of “simulation-meets-experiment” types of projects but no systematic effort to establish the rigor and the requirements on UQ and V&V that the cited reports have suggested are needed.

There are a few exceptions that are called out by the panel — for example in the work of SciDAC and the DOD Defense Modeling and Simulation Office, and theoretical work in Germany, Switzerland, and Austria — but these efforts are characterized as limited in scope and impact with respect to the size of the problem that needs to ultimately be addressed. Clearly, more understanding of the limits of the answers our computers are giving us is needed. This is especially important as we rely on them increasingly for answers about questions related to our own health and safety, and the welfare of our entire planet.

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