Meet the Exascale Apps

By Gary Johnson

April 12, 2012

In what will be a three-decade span between gigascale and exascale computing, HPC capability will have increased by a factor of one billion, but the apps that are projected to use this enormous increase in capability look pretty much like the gigascale ones. Are we missing opportunities as we push the apex of HPC higher?

Gigascale to Terascale

In February of 1991, the Office of Science and Technology Policy released the first “Blue Book” supplement to the President’s FY 1992 Budget Request for the new High Performance Computing and Communications Program. It was entitled “Grand Challenges: High Performance Computing and Communications” and contained a listing of the computational science and engineering challenges then seen as drivers for federal expenditures on HPC. Figure 2 from that report is reproduced below.

 Petascale to Exascale

In preparation for the current attempt to secure federal funding for exascale computing, the Department of Energy conducted a series of workshops entitled “Scientific Grand Challenges Workshop Series”. While this series only focused on science and engineering areas of importance to DOE’s mission, that mission is broad enough to view the grand challenges discussed there as typical of the applications areas foreseen as drivers for the move to exascale.

With the use of a bit of poetic license to prevent the reader’s eyes from glazing over, the table below attempts to convey the general character to these early 1990s gigascale to terascale applications and the exascale applications considered for the 2018-2025 timeframe (depending on whose guess about the arrival of exascale computing one chooses).

We see that over a span of 28 to 35 years, depending on how you count, the applications list remains substantially the same. A few of the 90s applications have dropped off the list – either through success or loss of interest. A couple of well-established applications: Nuclear Physics and Nuclear Energy Systems have been added in response to renewed interest in nuclear energy. To be sure, the other areas listed – the ones surviving multiple decades – have grown in complexity and broadened in applicability. What seems to be missing is the addition of any fundamentally new applications.

Over the decades since the publication of that first Blue Book, “apexscale” HPC has grown in capability by a factor of 1,000,000. In another decade, when exascale machines occupy the apex, they will be a factor of 1,000,000,000 more capable than those early 90s machines. Certainly, this enormous increase must present the opportunity to do a few fundamentally new things.

Capability Computing Usage Modes

In general, as HPC grows in capability, it can be used in three distinct ways:

  • Do what we’re currently doing, but faster or cheaper;
  • Undertake the logical extension of what we’re currently doing to use additional computing capabilities; or
  • Use the new and vastly more capable resource to do something we hadn’t seriously considered trying before.

Clearly and justifiably, we are using apexscale HPC in the first two ways. But what about the third? Have we run out of new ideas? Certainly not. But getting new apps on the agenda seems to have been either remarkably hard or of surprisingly little interest.

Exascale Readiness

Whether any new application candidate is, from inception, “exascale ready” seems considerably less important than its potential scalability. We are, after all, living in an age of scalable computing. Observe that many of the gigascale apps of the early 90s have readily survived, and thrived on, the transition to petascale and (soon) exascale. Did we coincidentally choose the complete collection of applications with this sort of potential for scalability back then or could there be others lurking in the wings?

Opportunities

Thinking of what we hadn’t thought of is always difficult and fraught with peril (you don’t know what you don’t know). However, the commercial and open science worlds have provided us with a few possibilities.

Big Data

Although several federally-funded applications areas have well-established needs for data crunching (e.g., high-energy physics, bioinformatics, and national security), the current opportunity in “Big Data” comes from the commercial world. Think: Social Data Analysis, Personal Analytics, Biobank, the Quantified Self, 23andMe, Healthrageous, Integrated Personal Omics, MyLifeBits. These are probably just the tip of the big data iceberg.

IBM has already launched Watson, with (beyond Jeopardy) foci on health care and financial services. Cray and Sandia National Laboratories have started a Supercomputing Institute for Learning and Knowledge Systems. NeuStar and the University of Illinois Urbana-Champaign have created a Big Data Research Facility. The federal government is also getting onboard with its recently announced Big Data Initiative. In fact, it’s interesting to note that the “Blue Book” accompanying the President’s FY 2013 budget request is strongly focused on big data and not the grand challenges of earlier blue books.

So, Big Data is probably a “no brainer” for the new applications category. Some of it may not be exascale yet, but there’s lots of room to grow.

Brain in a Box

This new application candidate has been advocated by Henry Markram at the Swiss Federal Institute of Technology in Lausanne (EFPL). Its official title is the Human Brain Project (HBP).

As described in a recent Nature article, it’s “an effort to build a supercomputer simulation that integrates everything known about the human brain, from the structures of ion channels in neural cell membranes up to mechanisms behind conscious decision-making.” Markram’s precursor Blue Brain Project at EFPL estimates that this is an exascale application (see figure below).

IBM is also a player in the activity, with its cognitive computing project called Systems of Neuromorphic Adaptive Plastic Scalable Electronics (SyNAPSE). This project claims that “By reproducing the structure and architecture of the brain—the way its elements receive sensory input, connect to each other, adapt these connections, and transmit motor output—the SyNAPSE project models computing systems that emulate the brain’s computing efficiency, size and power usage without being programmed.”

Thus, some form of simulation of the complete human brain seems like a keeper for our new applications short list.

Global-scale Systems

Under this heading, a couple of systems immediately come to mind: the global energy system and the global social system. Each seems worthy of a modeling effort.

In this vein, the European Commission has recently funded a “Big Science” pilot project, called FutureICT, “to understand and manage complex, global, socially interactive systems, with a focus on sustainability and resilience.” FutureICT intends to accomplish these goals “by developing new scientific approaches and combining these with the best established methods in areas like multi-scale computer modeling, social supercomputing, large-scale data mining and participatory platforms.” Sounds like there’s potential for an exascale application here.

To the best of our knowledge, there is no current effort to simulate the complete global energy system. However, given the critical nature of energy, from resource discovery and recovery, through transportation of energy materials, to production and distribution of energy, and disposition of by-products, it seems like having one or more full-scale, high- fidelity simulation tools on hand might be a good idea. Perhaps this will be part of the FutureICT project.

The Whole Planet

Thanks to a concerted international effort spanning a couple of decades, we now have some pretty good global climate models. This community effort has also set a shining example for “team science.”

Lately, the climate modeling community has begun using the term “Earth systems science,” as more phenomenology is added to the basic coupled ocean-atmosphere simulations. Laudable and valuable as these efforts may be, they still leave most of the planet out of the models. So, maybe we should model the whole planet.

The opportunity for such a whole planet model is made visible when one looks at the imagery of our Blue Marble. One immediately notices how thin the shell of the atmosphere is in comparison to the dimensions of our planet. The Earth’s volumetric mean radius is 6371 km. Current climate models reach about 30 km above the surface. The deepest point any ocean model needs to reach is about 12 km below the surface. So, our current modeling efforts are focused on a shell that is, at best, about 0.66 percent of the Earth’s radius. This shell represents about 1.96 percent of the Earth’s volume and 0.02 percent of its mass.

Note that the sort of whole planet model proposed here represents an extreme example of a multi-physics, multi-scale problem. The relevant temporal and spatial scales range from sub-millisecond molecular interactions to multi-millennia ice sheet models to million cubic kilometer modeling of the ionosphere.

The advantages of a fully integrated whole planet model are readily apparent and include applications for:

  • Disaster management and mitigation
  • Energy exploitation
  • Minerals exploration and recovery
  • Siting of critical facilities (e.g., nuclear power plants and waste repositories)
  • Understanding the impact of climate change on built infrastructure
  • Understanding the interactions among human, ecological and physical systems

The availability of such models would also serve to advance fundamental scientific understanding of our planet and its dynamics. Furthermore, undertaking to build such models would provide researchers in all of the relevant disciplines with a clear context for thinking about their research activities and how they contribute to the overall planet modeling effort.

Since the earth system models already in development will require trans-petascale computing capabilities, it is clear that exascale capability will be a bare minimum requirement for whole planet models.

The idea of building the sort of top-down whole planet model suggested here has also occurred to others. See, for example, the agenda of the Geneva-based International Centre for Earth Simulation (ICES). Furthermore, no discussion of this topic would be complete without paying homage to the ground-breaking efforts of Japan’s Earth Simulator Center.

Thinking outside the box

Making the case for new applications is a game that anyone can play. Here we have attempted to make the point that there may be worthwhile candidates lurking out there, beyond the view of our current exascale effort and its list of drivers.

If you don’t like these examples, please feel free to critique and improve them. If you have additional applications candidates, please make them known. The more frank and constructive discussion we have on this topic, the better and richer the future of HPC will be.

About the author

Gary M. Johnson is the founder of Computational Science Solutions, LLC, whose mission is to develop, advocate, and implement solutions for the global computational science and engineering community.

Dr. Johnson specializes in management of high performance computing, applied mathematics, and computational science research activities; advocacy, development, and management of high performance computing centers; development of national science and technology policy; and creation of education and research programs in computational engineering and science.

He has worked in Academia, Industry and Government. He has held full professorships at Colorado State University and George Mason University, been a researcher at United Technologies Research Center, and worked for the Department of Defense, NASA, and the Department of Energy.

He is a graduate of the U.S. Air Force Academy; holds advanced degrees from Caltech and the von Karman Institute; and has a Ph.D. in applied sciences from the University of Brussels.

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