THE VIEW FROM NPACI: AN INTERVIEW WITH SID KARIN

November 9, 2000

by Alan Beck, editor in chief LIVEwire

Dallas, Texas — Following is an interview with Sidney Karin, director of the National Partnership For Advanced Computational Infrastructure and professor at the San Diego Supercomputer Center, Department of Computer Science and Engineering, University of California, San Diego:

HPCwire. A casual perusal of research conducted at NPACI and NCSA appears to indicate that the former often concentrates upon on traditional supercomputing architectures while the latter tends to emphasize clustered COTS technology. Is this perception accurate? How have these two great NSF-funded consortia come to differentiate themselves? Should the pair now be consolidated?

KARIN: NPACI and the Alliance have a strong, collaborative relationship as part of the 3-year-old PACI program to build the nation’s next-generation of computational infrastructure, and during the next seven years of the program we’re looking forward to deploying new technologies that will foster even more important scientific advances than we’ve already accomplished.

As you know, in the fast-paced world of HPC there are no single best solutions for all time, and I think having the diversity of the twin partnership helps stimulate innovation and rapid progress.

In addition to a healthy competition where it’s appropriate, we also try to cooperate to avoid duplicating effort, and this arrangement allows each partnership to focus on its strengths. In particular, NPACI provides leadership in important technology areas such as providing the nation’s first teraflops-scale academic-use supercomputer, the most advanced data-intensive computing technologies, the largest storage archives, and the integration of these components. NPACI also provides leadership in a number of application areas including computational biology and bio-informatics.

In the NPACI vision, researchers collect data from observation, experiment and digital libraries, analyze the data with models run on supercomputers on the grid, visualize and share those data over the Web, and publish the results for the scientific community in digital libraries. As an example of how the NPACI vision has affected an entire scientific discipline, part of the inspiration for the National Virtual Observatory project, which has been endorsed by the astronomy community, has been NPACI’s Digital Sky project.

It’s also important to remember that HPC is more than hardware — it’s collaborations among people and software for integration and such essential technologies as data-handling. One key to NPACI’s success is that we’ve been able to bring together both computer scientists and discipline scientists in fields from bioinformatics and environmental informatics to astronomy. As a result of these collaborations, we’ve been able to help make important scientific advances and simultaneously deploy new computational technologies.

HPCwire: How would you characterize the state of US public policy vis-a-vis leading-edge HPC research? Are funding priorities sufficient; are they appropriately targeted? Is there an adequate understanding of HPC at the executive and legislative levels? What changes, if any, would you make?

KARIN: You have to keep your eye on the end result. There’s an overwhelming track record that investment in basic research, including HPC, that now produces a tremendous return on investment. The whole Internet and information technology revolution that’s happening right now is the result of basic research seeded decades ago. So to keep the economic engine — the engine of innovation — running takes timely investment, and it’s always useful to help policy makers remember that.

Scientific literacy is a serious problem in our society, and not just in HPC but in all areas of science and technology. So it’s incumbent upon us who are involved in HPC to maintain our efforts to educate policy makers and the legislature, which at NPACI we try to do through the 14,000 subscribers to our scientific journal EnVision and visitors to our Web publication Online. And of course we encourage everyone to stop by our booth at SC2000 and pick up free copies of our publications to see in more detail the amazing science and computational infrastructure that’s now emerging.

Specifically, about HPC I think policy makers often don’t realize that there’s a lot more than hardware involved. The most important trend that we see both now and into the future is not just the continued development of more powerful computers — whether tightly coupled or not — scientific instrumentation, extreme networks, or large data management and analysis facilities, but the critical need for people and software to integrate these components. And not only the integration into what’s called grid computing, but also the integration of various disciplines into bringing this whole system into reality. It’s this integrated, holistic, end-to-end system that will make possible tomorrow’s scientific advances.

HPCwire: What are the most important challenges facing HPC development as it enters the 21st century? How should members of our community best position themselves to deal with such issues?

KARIN: I think one of the most important challenges facing HPC development and the community is to communicate its own significance to the scientific enterprise. Building larger computers is, of course, vital to doing more advanced scientific computations, but to best serve the scientific community we have to treat the end-to-end problem, from collecting observations, moving data, performing computations, visualizing and interacting with results, and then repeating the process with further observation. We need to make these capabilities more widely available and transparent to the scientists, so they can focus less on computing and more on the science.

NPACI was established to create a computational infrastructure, and we realized that doing so requires a focus on the end-to-end problem, and the partnership brings together the country’s leading computer scientists and computational scientists for this purpose. NPACI has established major efforts to develop an end-to-end bioinformatics infrastructure for the vast amounts of biological data coming on-line, telescience environments for integrating remote control of scientific instruments in the computational neuroscience process and other disciplines, and interaction environments for managing and visualizing the massive data sets produced by scientific research ranging from neuroscience to oceanographic simulations, just to cite a few highlights.

Of course, the HPC community also has a responsibility toward educating both the nation’s policy makers and future generations of scientists about these opportunities, as I’ve already discussed.

HPCwire: How viable will quantum computing be within the next five years? Are there any other technological developments on the horizon that promise to revolutionize the face of HPC? What role is NPACI positioned to play in these areas?

KARIN: Quantum computing is one of many approaches being pursued to overcome the material limitations of traditional silicon-based chips. Innovative engineers will continue to merge the production of silicon-based chips with nanodevices, superconductors, quantum computing, biological molecules, and other technologies as each matures to production quality.

It’s important to remember, though, that it’s not just changes in the materials or construction of the processing circuits that will advance the hardware. The Cray (nee Tera) MTA in its first installation at SDSC shows that innovative architectural designs can advance high-end computing, and future systems will certainly employ such multi-threaded capabilities as well, witness the HTMT effort and the IBM on the Blue Gene system with which NPACI is collaborating.

I think that NPACI is well-positioned to play a central role in assembling the scientific expertise needed for integrating the most advanced hardware and software technologies into end-to-end scientific environments. Advances in HPC will hinge on areas such as data integration, network engineering, grid-enabled applications, and software that can take advantage of new and complex architectures. NPACI researchers are smoothing the transition of commodity clusters from experimental architectures into production environments with the recent release of the NPACI Rocks package. We’re uniting many types of systems into a seamless grid with the recent release of SDSC GridPort and PACI Genie. We, together with NCSA, are developing the software and hardware infrastructure that will be critical for tomorrow’s scientific discoveries on petaflops computing platforms with petabyte data sets moving across gigibit per second networks. Thus we are integrating new hardware technologies as well as computer and computational scientists, application and system software, into viable end-to-end solutions for scientific research.

HPCwire: Recently, several commercial enterprises have emerged that create a supercomputing environment by utilizing otherwise-idle processor time of Internet-connected PCs. What will be the most important factors determining the success or failure of such ventures?

KARIN: I don’t think it’s possible to predict who will succeed or fail in these ventures, because there are lots of factors involved, many of them by no means technological. These loosely coupled systems will work extremely well for certain classes of problems, and NPACI is, in fact, exploring the potential of such approaches with some of the leaders in the field. In the same way, grids and PC clusters are also important approaches to computing certain problems, and people now associated with NPACI led some of the early efforts in these areas. I personally feel that at least one of these approaches will be successful. However, you have to match the problem to the right systems. Many important scientific simulations will always run better — both in price-performance and time to solution — on single, tightly couple systems like Blue Horizon at SDSC. Others will achieve unprecedented performance levels on distributed megacomputers.

You have to picture scientific computing in context. We’ve already talked about the need for end-to-end solutions, but focusing just on computing power for a moment, there are many vitally important scientific problems that, even on the nation’s most powerful academic-use supercomputer, Blue Horizon — just upgraded to 1.7 teraflops — are barely approachable.

To support the spectrum of disciplines and the many classes of scientific problems, NPACI and all the PACI partnerships maintain computing environments with a variety of computing platforms. As technologies progress and new avenues of scientific interest emerge, NPACI and the PACI program will keep providing and exploring computing capabilities based on the most suitable technologies for the whole range of scientific disciplines. In addition to tightly coupled systems I expect distributed megacomputers to be part of our arsenal.

============================================================

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