ISC Keynote: Thomas Sterling’s Take on Whither HPC

By John Russell

June 20, 2019

Entertaining, insightful, and unafraid to launch the occasional verbal ICBM, HPC pioneer Thomas Sterling delivered his 16th annual closing keynote at ISC yesterday. He explored, among other things: exascale machinations; quantum’s bubbling money pot; Arm’s new HPC viability; Europe’s suspect plans (at least by Sterling) for a home-grown processor; HPE and Nvidia’s gobbling up of Cray and Mellanox; and even provided a tongue-in-cheek machine learning tutorial.

Thomas Sterling, ISC 2019 (photo credit: ISC)

Indeed, Sterling’s whirlwind sweep through the HPC landscape at ISC is always an adventure that prompts much laughter and a few cringes.

  • On the delayed Aurora project. “The Department of Energy in its infinite wisdom, had a very good team, Intel and Cray and other contributors, [and] did the obvious and right thing when the Aurora project failed. They gave the same people a much bigger budget with a much bigger objective, 1 exaflops, and in shorter time. That’s a winning formula for me.”
  • On China’s exascale plans. “My intent was to tell you China had selected its first exascale machine [and] it was going to be the Tianhe 3 and that this was a picture of the prototype,” said Sterling showing a slide. He learned otherwise at a very recent meeting with computer scientists from Asia. “What’s wrong with this is that it is all wrong. First, China has not decided what its number one machine will be; two and three, they said this is not a Tianhe and indeed [it’s] a Sunway machine. But it’s a nice picture.”

Obviously he had more to say about China’s ambitions but his willingness to absorb as well as to deliver barbs has proven to be a winning approach over the years, often saying publicly what others are murmuring behind closed doors.

Sterling made his presentation while seated, still recovering from recent surgery, but not lacking in zest. Sterling is best known as the “father of Beowulf” for his pioneering research in commodity/Linux cluster computing for which he shared the Gordon Bell Prize in 1997. Currently he professor of intelligent systems engineering at the Indiana University (IU) School of Informatics, Computing, and Engineering and also serves as the PI of the Continuum Computing Architecture Project at the Department of Intelligence Systems Engineering.

Presented here are a few of his slides and comments. Let’s start with the race to exascale which has dominated the high-end of HPC the last few years. While regional rivalries and the desire for competitive advantage are often used to characterize the race, Sterling said, “In my opinion this is about the expansion of the field and internationalizing of the ability of HPC for science and engineering and industry and commerce. [It’s] not about a competition.”

The global rising tide of powerful systems will raise all boats, he suggests. China may be the furthest along though the precise timetable for standing up an exaflops machine is unclear. NUDT has built a prototype system. “[It has] 512 nodes with the matrix 2000+ socket chip and this will be replaced by the matrix 3000 in preparation for whenever it delivers and exascale machine. It is an engineering wonder and really explains why the Asian community is doing so well in this field,” said Sterling.

Japan’s big recent news is around Fugaku (previously the post-K machine) which is based on Arm technology. “[It is] using Arm but only in the instruction set architecture. The actual core design is unique and new to Japan and Fujitsu and is combined with an integrated accelerator, the SVE, which is 512 bits wide,” said Sterling.

“Ever since the Earth Simulator, which of the gray beards here will remember, was stood up and was about 40 teraflops and horrified those of us in the west that suddenly woke up and realized we didn’t own the entire technology space. Jack Dongarra was quoted (in the New York Times) calling this “Computnik.” [Ever since then] the Japanese have constantly developed and deployed truly balanced architectures making it a little overkill with the communication and the bandwidth and low latency enhanced at expense for better balanced machines. [Fugaku] will be delivered midway 2021,” he said.

Fugaku will have a TOFU-D 6D toroidal mesh. “I hate Tofu (the food),” said Sterling as an aside, but in this case “it stands for toroidal fusion, and once again confers a tremendous amount of integral bandwidth to the future machine. This will be a major contributor to the field.”

He had praise and disappointment for the U.S. effort to reach exascale computing. The Exascale Computing Project (ECP) focused on developing needed software technology and adapting applications got some love from Sterling. So did recent leadership-class, pre-exascale machines, Summit and Sierra. Not so much for Aurora, being built by Intel and Cray, which has been delayed and is now scheduled to be the first U.S. exascale machine (possibly; Frontier at Oak Ridge has a competitive timeline).

So much has already been said about Aurora. Sterling glibly noted it was DOA but quickly added machines sometimes fail and that’s part of progress and not to be excessively dwelled upon. Not much is known about the forthcoming Aurora. It will leverage Cray’s new Shasta design and Slingshot technology; use and Intel CPU and an Intel GPU about which very little is known; and be deployed at Argonne National Laboratory.

Tongue-in-cheek, Sterling said, “It will be truly leading edge one way or the other, [but] I don’t know if it will be the back edge or the front edge. I should be giving credit to Intel and Cray…Xeons will be there. Some will be fat, some will be skinny, and then there’s a three-letter acronym that will be included. As far as I could tell this will not be anything like the original Aurora machine but we will all find out in a little while. This is a general purpose machine; that’s very important, for simulation, streaming data analytics, machine intelligence.”

He notes some ideas about what might be in Aurora eventually and indicated they seemed reasonable.

Sterling acknowledged Europe’s steady progress in building out its HPC infrastructure and relative success working closely with industry. He also acknowledged it’s in a bit of bind: “The truth is the European community has a very important problem to solve. This is [that] they are a consumer of about 30 percent of the worldwide HPC yet they bring in only about three percent of the [revenue]; this is clearly an unbalanced business model and has to be addressed and resolved.”

Sterling wasn’t optimistic about current plans to build a home-grown advanced processor. It is a “goal for which they are ill prepared and that is not only to develop a new processor for HPC and exascale but rather to do so with an experience base with which they have developed no such processors in the last ten years,” he said.

One take home idea, he suggested, was that HPC is splitting into two distinct communities that serve different purposes: one advances the state of the art in technology and the other doing the science and engineering in industry, academia, and commerce using that technology.

It’s not surprising quantum computing got so much attention. As he pointed out there is a significant amount of government and private money pouring into quantum computing, a la the U.S. $1.25B Quantum Initiative.

“Here’s an area that is, frankly, really speculative, and yet the world has constructively and appropriately come together to make a bet, to project its hope for the future,” he said. It’s probably safe to infer he wonders a bit about how quickly quantum computing will become a practical tool. That said, there’s a global race for QC afoot and what seems like a sudden gush of funding.

Right now it’s not clear what the winning qubit technology will be. Semiconductor-based superconducting technology is the dominant track but others are emerging. Much of the ongoing work isn’t on qubit technology but on software – the algorithms and programming environment models. Many of those most active QC companies, he noted, are making their technology available by the web.

“[It’s] truly an outreach to users, partly to educate us and partly to provide feedback to them about where the field has to evolve,” said Sterling. He poked a little fun at the notions of quantum supremacy and quantum dominance as marketing driven.

“The Japanese are also investing significantly in QC,” said Sterling. “The second bullet (slide below) says they are slow to bring this to fruition. This is true but it also might be the right strategy. The Japanese involving major industry partners across Japan which are in the search phase, not rushing too much to jump on any particular [technology] but continuing to evolve their perception of the possibilities, the most promising paths. So it may be the slow and steady wins the race here. We can only wait and see.”

Least is known about China’s quantum computing efforts. It is planning a massive single center in which to drive its quantum efforts. “The center for quantum science should be open about a year from now. There is already significant work with the Chinese Academy of Science,” he said. It will be interesting to watch how quickly China’s centralized efforts progress versus the U.S.’s very diverse efforts encompassing academia and private enterprise.

Capturing the full scope of Sterling’s comments is beyond this short article. ISC will likely make the recorded version available. He spent some time paying tribute to recent HPC award winners as he usually does.

In summary, Sterling is fond of picking an event in science that was enabled by HPC and marks a significant breakthrough. This year he chose the imaging of a black hole at the center of NGC 4486, a galaxy much larger than the Milky Way and with five times as many stars.

“Black holes, outside of science fiction, are confusing, badly modeled, and there is still some reticence to fully embrace it. Until now. A very large collaboration involving both HPC and eight major microwave antennas has been collecting data for more than a year and processing that data,” he noted. The results, of course, were stunning images of the event horizon around the black hole.

On to ISC2020

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industy updates delivered to you every week!

The Case for an Edge-Driven Future for Supercomputing

September 24, 2021

“Exascale only becomes valuable when it’s creating and using data that we care about,” said Pete Beckman, co-director of the Northwestern-Argonne Institute of Science and Engineering (NAISE), at the most recent HPC Read more…

Three Universities Team for NSF-Funded ‘ACES’ Reconfigurable Supercomputer Prototype

September 23, 2021

As Moore’s law slows, HPC developers are increasingly looking for speed gains in specialized code and specialized hardware – but this specialization, in turn, can make testing and deploying code trickier than ever. Now, researchers from Texas A&M University, the University of Illinois at Urbana... Read more…

Qubit Stream: Monte Carlo Advance, Infosys Joins the Fray, D-Wave Meeting Plans, and More

September 23, 2021

It seems the stream of quantum computing reports never ceases. This week – IonQ and Goldman Sachs tackle Monte Carlo on quantum hardware, Cambridge Quantum pushes chemistry calculations forward, D-Wave prepares for its Read more…

Asetek Announces It Is Exiting HPC to Protect Future Profitability

September 22, 2021

Liquid cooling specialist Asetek, well-known in HPC circles for its direct-to-chip cooling technology that is inside some of the fastest supercomputers in the world, announced today that it is exiting the HPC space amid multiple supply chain issues related to the pandemic. Although pandemic supply chain... Read more…

TACC Supercomputer Delves Into Protein Interactions

September 22, 2021

Adenosine triphosphate (ATP) is a compound used to funnel energy from mitochondria to other parts of the cell, enabling energy-driven functions like muscle contractions. For ATP to flow, though, the interaction between the hexokinase-II (HKII) enzyme and the proteins found in a specific channel on the mitochondria’s outer membrane. Now, simulations conducted on supercomputers at the Texas Advanced Computing Center (TACC) have simulated... Read more…

AWS Solution Channel

Introducing AWS ParallelCluster 3

Running HPC workloads, like computational fluid dynamics (CFD), molecular dynamics, or weather forecasting typically involves a lot of moving parts. You need a hundreds or thousands of compute cores, a job scheduler for keeping them fed, a shared file system that’s tuned for throughput or IOPS (or both), loads of libraries, a fast network, and a head node to make sense of all this. Read more…

The Latest MLPerf Inference Results: Nvidia GPUs Hold Sway but Here Come CPUs and Intel

September 22, 2021

The latest round of MLPerf inference benchmark (v 1.1) results was released today and Nvidia again dominated, sweeping the top spots in the closed (apples-to-apples) datacenter and edge categories. Perhaps more interesti Read more…

The Case for an Edge-Driven Future for Supercomputing

September 24, 2021

“Exascale only becomes valuable when it’s creating and using data that we care about,” said Pete Beckman, co-director of the Northwestern-Argonne Institut Read more…

Three Universities Team for NSF-Funded ‘ACES’ Reconfigurable Supercomputer Prototype

September 23, 2021

As Moore’s law slows, HPC developers are increasingly looking for speed gains in specialized code and specialized hardware – but this specialization, in turn, can make testing and deploying code trickier than ever. Now, researchers from Texas A&M University, the University of Illinois at Urbana... Read more…

Qubit Stream: Monte Carlo Advance, Infosys Joins the Fray, D-Wave Meeting Plans, and More

September 23, 2021

It seems the stream of quantum computing reports never ceases. This week – IonQ and Goldman Sachs tackle Monte Carlo on quantum hardware, Cambridge Quantum pu Read more…

Asetek Announces It Is Exiting HPC to Protect Future Profitability

September 22, 2021

Liquid cooling specialist Asetek, well-known in HPC circles for its direct-to-chip cooling technology that is inside some of the fastest supercomputers in the world, announced today that it is exiting the HPC space amid multiple supply chain issues related to the pandemic. Although pandemic supply chain... Read more…

TACC Supercomputer Delves Into Protein Interactions

September 22, 2021

Adenosine triphosphate (ATP) is a compound used to funnel energy from mitochondria to other parts of the cell, enabling energy-driven functions like muscle contractions. For ATP to flow, though, the interaction between the hexokinase-II (HKII) enzyme and the proteins found in a specific channel on the mitochondria’s outer membrane. Now, simulations conducted on supercomputers at the Texas Advanced Computing Center (TACC) have simulated... Read more…

The Latest MLPerf Inference Results: Nvidia GPUs Hold Sway but Here Come CPUs and Intel

September 22, 2021

The latest round of MLPerf inference benchmark (v 1.1) results was released today and Nvidia again dominated, sweeping the top spots in the closed (apples-to-ap Read more…

Why HPC Storage Matters More Now Than Ever: Analyst Q&A

September 17, 2021

With soaring data volumes and insatiable computing driving nearly every facet of economic, social and scientific progress, data storage is seizing the spotlight. Hyperion Research analyst and noted storage expert Mark Nossokoff looks at key storage trends in the context of the evolving HPC (and AI) landscape... Read more…

GigaIO Gets $14.7M in Series B Funding to Expand Its Composable Fabric Technology to Customers

September 16, 2021

Just before the COVID-19 pandemic began in March 2020, GigaIO introduced its Universal Composable Fabric technology, which allows enterprises to bring together Read more…

Ahead of ‘Dojo,’ Tesla Reveals Its Massive Precursor Supercomputer

June 22, 2021

In spring 2019, Tesla made cryptic reference to a project called Dojo, a “super-powerful training computer” for video data processing. Then, in summer 2020, Tesla CEO Elon Musk tweeted: “Tesla is developing a [neural network] training computer called Dojo to process truly vast amounts of video data. It’s a beast! … A truly useful exaflop at de facto FP32.” Read more…

Enter Dojo: Tesla Reveals Design for Modular Supercomputer & D1 Chip

August 20, 2021

Two months ago, Tesla revealed a massive GPU cluster that it said was “roughly the number five supercomputer in the world,” and which was just a precursor to Tesla’s real supercomputing moonshot: the long-rumored, little-detailed Dojo system. “We’ve been scaling our neural network training compute dramatically over the last few years,” said Milan Kovac, Tesla’s director of autopilot engineering. Read more…

Esperanto, Silicon in Hand, Champions the Efficiency of Its 1,092-Core RISC-V Chip

August 27, 2021

Esperanto Technologies made waves last December when it announced ET-SoC-1, a new RISC-V-based chip aimed at machine learning that packed nearly 1,100 cores onto a package small enough to fit six times over on a single PCIe card. Now, Esperanto is back, silicon in-hand and taking aim... Read more…

CentOS Replacement Rocky Linux Is Now in GA and Under Independent Control

June 21, 2021

The Rocky Enterprise Software Foundation (RESF) is announcing the general availability of Rocky Linux, release 8.4, designed as a drop-in replacement for the soon-to-be discontinued CentOS. The GA release is launching six-and-a-half months after Red Hat deprecated its support for the widely popular, free CentOS server operating system. The Rocky Linux development effort... Read more…

Intel Completes LLVM Adoption; Will End Updates to Classic C/C++ Compilers in Future

August 10, 2021

Intel reported in a blog this week that its adoption of the open source LLVM architecture for Intel’s C/C++ compiler is complete. The transition is part of In Read more…

Hot Chips: Here Come the DPUs and IPUs from Arm, Nvidia and Intel

August 25, 2021

The emergence of data processing units (DPU) and infrastructure processing units (IPU) as potentially important pieces in cloud and datacenter architectures was Read more…

AMD-Xilinx Deal Gains UK, EU Approvals — China’s Decision Still Pending

July 1, 2021

AMD’s planned acquisition of FPGA maker Xilinx is now in the hands of Chinese regulators after needed antitrust approvals for the $35 billion deal were receiv Read more…

Google Launches TPU v4 AI Chips

May 20, 2021

Google CEO Sundar Pichai spoke for only one minute and 42 seconds about the company’s latest TPU v4 Tensor Processing Units during his keynote at the Google I Read more…

Leading Solution Providers

Contributors

HPE Wins $2B GreenLake HPC-as-a-Service Deal with NSA

September 1, 2021

In the heated, oft-contentious, government IT space, HPE has won a massive $2 billion contract to provide HPC and AI services to the United States’ National Security Agency (NSA). Following on the heels of the now-canceled $10 billion JEDI contract (reissued as JWCC) and a $10 billion... Read more…

10nm, 7nm, 5nm…. Should the Chip Nanometer Metric Be Replaced?

June 1, 2020

The biggest cool factor in server chips is the nanometer. AMD beating Intel to a CPU built on a 7nm process node* – with 5nm and 3nm on the way – has been i Read more…

Julia Update: Adoption Keeps Climbing; Is It a Python Challenger?

January 13, 2021

The rapid adoption of Julia, the open source, high level programing language with roots at MIT, shows no sign of slowing according to data from Julialang.org. I Read more…

Quantum Roundup: IBM, Rigetti, Phasecraft, Oxford QC, China, and More

July 13, 2021

IBM yesterday announced a proof for a quantum ML algorithm. A week ago, it unveiled a new topology for its quantum processors. Last Friday, the Technical Univer Read more…

Intel Launches 10nm ‘Ice Lake’ Datacenter CPU with Up to 40 Cores

April 6, 2021

The wait is over. Today Intel officially launched its 10nm datacenter CPU, the third-generation Intel Xeon Scalable processor, codenamed Ice Lake. With up to 40 Read more…

Frontier to Meet 20MW Exascale Power Target Set by DARPA in 2008

July 14, 2021

After more than a decade of planning, the United States’ first exascale computer, Frontier, is set to arrive at Oak Ridge National Laboratory (ORNL) later this year. Crossing this “1,000x” horizon required overcoming four major challenges: power demand, reliability, extreme parallelism and data movement. Read more…

Intel Unveils New Node Names; Sapphire Rapids Is Now an ‘Intel 7’ CPU

July 27, 2021

What's a preeminent chip company to do when its process node technology lags the competition by (roughly) one generation, but outmoded naming conventions make it seem like it's two nodes behind? For Intel, the response was to change how it refers to its nodes with the aim of better reflecting its positioning within the leadership semiconductor manufacturing space. Intel revealed its new node nomenclature, and... Read more…

Latest MLPerf Results: Nvidia Shines but Intel, Graphcore, Google Increase Their Presence

June 30, 2021

While Nvidia (again) dominated the latest round of MLPerf training benchmark results, the range of participants expanded. Notably, Google’s forthcoming TPU v4 Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire