SC13 Talk Pushes HPC in New Educational Directions

By Nicole Hemsoth

November 19, 2013

According to Dr. Thom Dunning from University of Illinois at Urbana-Champaign, science and engineering research has been revolutionized by computation but, to date computing has largely been used to organize, prepare for and disseminate courses. Dunning says that “The potential of using these technologies to teach students the fundamental principles of a subject through authentic computational simulation is largely unexplored.”

During a presentation today (Tuesday) as part of the SC13 Invited Speakers program, Dunning will discuss how “computational tools and simulations has been used to help teach high school and college chemistry. The simulations enable students to gain a deep, rich appreciation for the basic principles of chemistry. Further, the use of computational tools is enthusiastically embraced by teachers and students, results in improved performance of both, and leads to increased student interest in chemistry.”

Dunning’s talk is at 1:30 today at SC13 in Denver…a full interview is below that hits on the main topics he is set to discuss, plus a broader view of what tools and opportunities HPC pushes to other domains.

HPCwire: Can you give us a sense of what the various roles are for HPC in education—where does it fit in with scientific and broader educational goals?

Dunning: The overarching thesis I have is, as computational simulation has revolutionized scientific and engineering research, it can also revolutionize scientific and engineering education.

It’s very clear that computing and particular simulations had a major impact on most fields of science. It adds a new way for us to understand how the world operates, whether it’s the chemical world, the physical world, the biological world or some combination of all of them, which is of course what happens with global climate simulation. And we really have to think about how it we effectively use those kinds of simulations in education.

I’m a chemist by background, and all chemical phenomena are determined by molecules. But students never see molecules. It might at times almost sound to them like a fiction. But in fact they’re very real, and what you need to do for the student is to make them real. And that’s something we can do with computational simulation.

And in that case it could be as simple as allowing them to draw a molecule on a computer screen and manipulate it—turn it around, look at various sides of it, zoom in and out. Or it could be as sophisticated as taking a molecule and computing its vibrational spectra and then tying that in with global warming.

A teacher might ask why it is that the CO2 molecule and the buildup of CO2 in the atmosphere leads to global warming. To help students answer the question, the teacher could use computational simulation to show the students how the structure and the way the molecule moves is tied to the global warming phenomenon.

We have a project that has now been completed at NCSA for six years, called the Institute for Chemical Literacy through Computational Science. Although the name says “computational science,” computational simulation was really key. There we were looking at tools that would allow the teachers to work with students to visualize these kinds of molecular phenomena, and we saw a significant increase in not only teacher knowledge, but we also saw a significant increase in student knowledge as a result of this intervention.

HPCwire: So it sounds as though making science real and tangible then is one of the core components of what you’ll be talking about.

Dunning: Yes, and if you think about it, this is one of the issues that we have in science: many of the fundamental, underlying concepts from continental drift to the inner workings of the universe aren’t easily visualized.

But by having these simulations one can really make all of this very much more real for students and I think that’s the major benefit of computational simulation in the classroom. It really makes whatever the student is studying more real because they can interact with it and watch how it responds and thereby learn the behavior they’re supposed to be studying.

HPCwire: You’re a chemist by training, so is this the way you found your way into the high performance computing space?

Dunning: Yes, it certainly was a factor when I was an undergraduate. I went to a university that, when I was a sophomore, had just gotten a new computer: an IBM 1620. I was taking a course in quantum mechanics and suddenly it dawned on me that quantum mechanics was actually the basis for all of the chemistry that I had been taking. Then I started looking at what the equations looked like and I quickly realized I couldn’t solve them, and it took me a little while longer to realize that nobody could solve some of them. But with computers you could.

So that really got me interested in the use of computing to study various kinds of chemical research problems. And even for me it made it a lot more real—I was actually doing something. And I was manipulating chemistry at its most fundamental level: the molecular level.

When you go into the laboratory and you pour two solutions together, the outcome is being determined at that fundamental level, but that’s not what you see and that’s not what you experience. But when you actually do the computational simulations you can get a direct experience.

HPCwire: Speaking of that new generation, you’ve seen a number of trends come and go when it comes to computer scientists and students interested in HPC. Have you seen a dip or rise or stagnation around high performance computing as a career path? What ways are people coming into it? What trends have you seen over time?

Dunning: Well I guess what we’re seeing really is an increasing number of scientific communities that really need HPC. Either the models they use to describe the phenomena they’re interested in have gotten sufficiently complicated and sophisticated that now they need HPC to solve them, or on the data side communities now get to the point where the amount of data that they’re generating also requires HPC to be able to manage the data and to analyze it. So I think this is an area that is going to continue to grow over the next decade. And of course in chemistry it’s grown tremendously since I’ve been in the field.

I like to say that when I first started in the 1960s, if you had a conference that had a couple hundred theoretical and computational chemists in it, that was the world’s supply. Now there are thousands of theoretical and computational chemists—even tens of thousands. But even more important than that, even experimental chemists now routinely use computational simulation to gain insights into the experiments that they do. So it’s really expanded far beyond the folks who have traditionally done that type of work, to the point that I would say half of all chemists use computational simulations in one way or another.

HPCwire: So a lot of the interest in HPC then is being fed by the actual end-users of some of these applications? What about on the other side of it—people who are actively pursuing a career in high performance systems to help empower these systems. Are there some trends that you’ve noticed there?

Dunning: I don’t think I can be definitive there, but I can at least give you an impression. My impression is that when I started back in the 60s, most of the people that I ran into who were computer scientists were focused on scientific computing and what would today be called high performance computing, although it was CDC 6600s and things like that back then.

Now computing has gotten to be such a large, broad field that the number of computer scientists who are actually focused on high performance computing is now a pretty small fraction of the total field of computer science. And I would say that that area has certainly not seen the growth that computer science as a whole has seen because of exactly that.

HPCwire: Could you explain in more detail what some of those reasons may be?

Dunning: I think that for people who have an interest in computing, back in the 60s and 70s, scientific computing was the hot thing to do. In fact, it was the only thing you could do. Now if you’re interested in computing and you want to be a computer scientist, there’s a very broad field that you could be involved in: everything from the Web to real-time processing or embedded processing, and high performance computing is only a very small fraction of computer science at this point.

HPCwire: I see a lot of job postings and kind of “hipness” around the big data jobs and the label of “data scientist,” which is a way to call advanced computing something else completely. Maybe you just answered this question by what you just said, but do you want to comment on that transition into this culture of big data? There’s a piece of it in HPC, and a lot of these guys say, “Well, HPC has always been big data, it’s just data, and yes it’s gotten bigger.” There’s a little bit of push and pull between those two cultures, but there’s really nothing new.

Dunning: I think there’s a measure of truth to both of the comments. I think it’s correct for people in HPC to say, “We’ve been dealing with big data basically from the beginning. It’s been growing year by year.” And that’s true. On the other hand there are new sources of data now. There are sensor networks that are being put in place. All of the telescopes that are designed now basically are digital instruments—they are no longer analog instruments.

So there’s a huge explosion that’s occurring now both inside and outside of HPC. And I think that’s the big change that’s happened over the course of the past few years.

When I first came to NCSA in 2004 we were talking about the rapid increase in the growth of data. It took NCSA the first 19 years of its existence to archive a petabyte of data. By the twentieth year we had a second petabyte of data. And we’ve been adding substantial amounts of data into the archive ever since then.

But now we’re talking about instruments like the LSST, the Large Synoptic Survey Telescope. Once it’s in operation it will, by itself, generate something on the order of 100 to 200 petabytes of data. So in some ways we’re not quite there yet, but it’s coming at us pretty fast at this point. For these other sources of data are going to become as dominant if not more dominant than the amount of data that’s produced in high performance computing.

And I think that one of the things that’s going to happen here is that it’s not only going to revolutionize science and engineering, (the telescopes being a good example of how it revolutionizes science,) but it’s also going to revolutionize the behavioral, social and economic sciences also. Because as they get to the point that they start collecting the data, which is now so much easier to do because so much data is born digitally to begin with, they’re going to get new insights into the problems they’re studying that wouldn’t be possible to get otherwise.

HPCwire: It sounds like that brings us full circle back to the work that you’ll eventually be doing with the sustainable cities project, where you’re gathering sociological data plus sensor data plus GIS and other types of data, right?

Dunning: That’s correct. Looking at cities is putting all of that information together and really understanding what that data is telling you about how you can make the city operate more efficiently both energy-wise and people-wise.

HPCwire: Does this intersect at all with a project in the Middle East that Thomas Zachariah was doing regarding connected cities. Are these similar in nature?

Dunning: I’m not sure exactly what it is that Thomas is doing, but at NYU there’s a center called the Center for Urban Science and Policy and that’s one of its goal also: to use the city of New York as a testbed for understanding how you can improve the efficiency of cities and improve the quality of life in cities.

And a lot of this is drive by the fact that right now. I believe, in the U.S. about 50 percent of the population lives in cities and by the time we get to 2050 it’s going to be more like 70 percent. And this is a trend that’s happening worldwide, not just in the U.S So if you’re really going to focus on a major problem where you need to enhance energy efficiency and sustainability, cities are a natural testbed for that and one of the problems you really want to settle.

HPCwire: Is there any other takeaway from the talks you’re expected to give at SC that you want to share here and now?

Dunning: I guess my goal is to make this a call to arms—to really get the folks in HPC interested in talking to their colleagues in science and engineering.

Right now most of that is done in research, and I think it’s time for us to start that discussion on the teaching side. And I know here at Illinois I’ve had conversations with my colleagues on the faculty here and there is a lot of pent-up desire to introduce computational simulation into the undergraduate classroom because they know it could have a big impact. But it really takes a partnership between the folks in HPC and the faculty to be able to bring that about in each of the universities.

So I’m kind of hoping that people come away from my talk saying “You know, I need to go over and have a talk with Joe in Physics because I know he’s interested in computational simulation. I wonder what his thoughts are on using computational simulation in the classroom?”

I think it would just be terrific if you had a class in meteorology and the students each got to run their weather forecasts. In fact you change the initial conditions slightly so that all of a sudden when you say, “There’s a 30 percent chance of rain,” the students would really have a real understanding of what that meant because they could look at all the different simulations that all the other students ran to see in 30 percent of them there was rain in this area. It would make it really concrete as to what terms like that mean, which now people sort of having a feeling for, but not an in-depth understanding.

Same kind of thing with climate simulation. Having students run climate simulations and understand where some of the issues might be in some of the simulations and where the simulations are very stable. They’re all predicting very similar kinds of trends, which would give you much more confidence in that particular number. Then seeing where they aren’t producing things that are as consistent, so maybe that’s an area where the models need to be improved.

HPCwire: By the way, what are you up to now? We recently published news that you’ve retired as NCSA director…what next?

Dunning: Well, that doesn’t actually happen until December 31st, so I’m still officially the NCSA director. After that I will be joining Pacific Northwest National Laboratory and the University of Washington to help them set up a joint institute for advanced computing between the two institutions.

They’re really just formulating it at this point. This is an entirely new institute, so one of the things that we’ll be doing over the next couple of years is looking at areas of common interest between the institutions and looking for opportunities for them to collaborate on various projects in advanced computing.

And it will be both what I refer to as “scientific computing” and data analysis, which is the fundamental base for the activity, but it will also look at various kinds of applications. For example: exascale science and engineering, which involves looking at the science of cities to try to understand how they operate and how one can make them more sustainable and energy efficient.

But those specific areas have not yet been decided on—we’ll be holding workshops over the next couple of years to define what some of those major thrusts will be. And then there will be lots of individual collaborations between staff at Pacific Northwest National Laboratory and faculty at the University of Washington.

HPCwire: That doesn’t sound much like retirement.

Dunning: I really was planning on retiring. Although this is the second time I’ve tried to retire and as soon as I do people come knocking on my door, saying, “Hey, I’ve got something great for you to do.”

This one actually holds a special place for me in that when I was at Pacific Northwest National Laboratory one of my goals was, in fact, to get the two institutions, who are the two major research institutions in the state of Washington, working more closely together. So this is something that’s close to my heart.

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!

Weekly Twitter Roundup (Feb. 23, 2017)

February 23, 2017

Here at HPCwire, we aim to keep the HPC community apprised of the most relevant and interesting news items that get tweeted throughout the week. Read more…

By Thomas Ayres

HPE Server Shows Low Latency on STAC-N1 Test

February 22, 2017

The performance of trade and match servers can be a critical differentiator for financial trading houses. Read more…

By John Russell

HPC Financial Update (Feb. 2017)

February 22, 2017

In this recurring feature, we’ll provide you with financial highlights from companies in the HPC industry. Check back in regularly for an updated list with the most pertinent fiscal information. Read more…

By Thomas Ayres

Rethinking HPC Platforms for ‘Second Gen’ Applications

February 22, 2017

Just what constitutes HPC and how best to support it is a keen topic currently. Read more…

By John Russell

HPE Extreme Performance Solutions

O&G Companies Create Value with High Performance Remote Visualization

Today’s oil and gas (O&G) companies are striving to process datasets that have become not only tremendously large, but extremely complex. And the larger that data becomes, the harder it is to move and analyze it – particularly with a workforce that could be distributed between drilling sites, offshore rigs, and remote offices. Read more…

HPC Technique Propels Deep Learning at Scale

February 21, 2017

Researchers from Baidu’s Silicon Valley AI Lab (SVAIL) have adapted a well-known HPC communication technique to boost the speed and scale of their neural network training and now they are sharing their implementation with the larger deep learning community. Read more…

By Tiffany Trader

IDC: Will the Real Exascale Race Please Stand Up?

February 21, 2017

So the exascale race is on. And lots of organizations are in the pack. Government announcements from the US, China, India, Japan, and the EU indicate that they are working hard to make it happen – some sooner, some later. Read more…

By Bob Sorensen, IDC

ExxonMobil, NCSA, Cray Scale Reservoir Simulation to 700,000+ Processors

February 17, 2017

In a scaling breakthrough for oil and gas discovery, ExxonMobil geoscientists report they have harnessed the power of 717,000 processors – the equivalent of 22,000 32-processor computers – to run complex oil and gas reservoir simulation models. Read more…

By Doug Black

TSUBAME3.0 Points to Future HPE Pascal-NVLink-OPA Server

February 17, 2017

Since our initial coverage of the TSUBAME3.0 supercomputer yesterday, more details have come to light on this innovative project. Of particular interest is a new board design for NVLink-equipped Pascal P100 GPUs that will create another entrant to the space currently occupied by Nvidia's DGX-1 system, IBM's "Minsky" platform and the Supermicro SuperServer (1028GQ-TXR). Read more…

By Tiffany Trader

HPC Technique Propels Deep Learning at Scale

February 21, 2017

Researchers from Baidu’s Silicon Valley AI Lab (SVAIL) have adapted a well-known HPC communication technique to boost the speed and scale of their neural network training and now they are sharing their implementation with the larger deep learning community. Read more…

By Tiffany Trader

IDC: Will the Real Exascale Race Please Stand Up?

February 21, 2017

So the exascale race is on. And lots of organizations are in the pack. Government announcements from the US, China, India, Japan, and the EU indicate that they are working hard to make it happen – some sooner, some later. Read more…

By Bob Sorensen, IDC

TSUBAME3.0 Points to Future HPE Pascal-NVLink-OPA Server

February 17, 2017

Since our initial coverage of the TSUBAME3.0 supercomputer yesterday, more details have come to light on this innovative project. Of particular interest is a new board design for NVLink-equipped Pascal P100 GPUs that will create another entrant to the space currently occupied by Nvidia's DGX-1 system, IBM's "Minsky" platform and the Supermicro SuperServer (1028GQ-TXR). Read more…

By Tiffany Trader

Tokyo Tech’s TSUBAME3.0 Will Be First HPE-SGI Super

February 16, 2017

In a press event Friday afternoon local time in Japan, Tokyo Institute of Technology (Tokyo Tech) announced its plans for the TSUBAME3.0 supercomputer, which will be Japan’s “fastest AI supercomputer,” Read more…

By Tiffany Trader

Drug Developers Use Google Cloud HPC in the Fight Against ALS

February 16, 2017

Within the haystack of a lethal disease such as ALS (amyotrophic lateral sclerosis / Lou Gehrig’s Disease) there exists, somewhere, the needle that will pierce this therapy-resistant affliction. Read more…

By Doug Black

Azure Edges AWS in Linpack Benchmark Study

February 15, 2017

The “when will clouds be ready for HPC” question has ebbed and flowed for years. Read more…

By John Russell

Is Liquid Cooling Ready to Go Mainstream?

February 13, 2017

Lost in the frenzy of SC16 was a substantial rise in the number of vendors showing server oriented liquid cooling technologies. Three decades ago liquid cooling was pretty much the exclusive realm of the Cray-2 and IBM mainframe class products. That’s changing. We are now seeing an emergence of x86 class server products with exotic plumbing technology ranging from Direct-to-Chip to servers and storage completely immersed in a dielectric fluid. Read more…

By Steve Campbell

Cray Posts Best-Ever Quarter, Visibility Still Limited

February 10, 2017

On its Wednesday earnings call, Cray announced the largest revenue quarter in the company’s history and the second-highest revenue year. Read more…

By Tiffany Trader

For IBM/OpenPOWER: Success in 2017 = (Volume) Sales

January 11, 2017

To a large degree IBM and the OpenPOWER Foundation have done what they said they would – assembling a substantial and growing ecosystem and bringing Power-based products to market, all in about three years. Read more…

By John Russell

US, China Vie for Supercomputing Supremacy

November 14, 2016

The 48th edition of the TOP500 list is fresh off the presses and while there is no new number one system, as previously teased by China, there are a number of notable entrants from the US and around the world and significant trends to report on. Read more…

By Tiffany Trader

Lighting up Aurora: Behind the Scenes at the Creation of the DOE’s Upcoming 200 Petaflops Supercomputer

December 1, 2016

In April 2015, U.S. Department of Energy Undersecretary Franklin Orr announced that Intel would be the prime contractor for Aurora: Read more…

By Jan Rowell

D-Wave SC16 Update: What’s Bo Ewald Saying These Days

November 18, 2016

Tucked in a back section of the SC16 exhibit hall, quantum computing pioneer D-Wave has been talking up its new 2000-qubit processor announced in September. Forget for a moment the criticism sometimes aimed at D-Wave. This small Canadian company has sold several machines including, for example, ones to Lockheed and NASA, and has worked with Google on mapping machine learning problems to quantum computing. In July Los Alamos National Laboratory took possession of a 1000-quibit D-Wave 2X system that LANL ordered a year ago around the time of SC15. Read more…

By John Russell

Enlisting Deep Learning in the War on Cancer

December 7, 2016

Sometime in Q2 2017 the first ‘results’ of the Joint Design of Advanced Computing Solutions for Cancer (JDACS4C) will become publicly available according to Rick Stevens. He leads one of three JDACS4C pilot projects pressing deep learning (DL) into service in the War on Cancer. Read more…

By John Russell

IBM Wants to be “Red Hat” of Deep Learning

January 26, 2017

IBM today announced the addition of TensorFlow and Chainer deep learning frameworks to its PowerAI suite of deep learning tools, which already includes popular offerings such as Caffe, Theano, and Torch. Read more…

By John Russell

HPC Startup Advances Auto-Parallelization’s Promise

January 23, 2017

The shift from single core to multicore hardware has made finding parallelism in codes more important than ever, but that hasn’t made the task of parallel programming any easier. Read more…

By Tiffany Trader

CPU Benchmarking: Haswell Versus POWER8

June 2, 2015

With OpenPOWER activity ramping up and IBM’s prominent role in the upcoming DOE machines Summit and Sierra, it’s a good time to look at how the IBM POWER CPU stacks up against the x86 Xeon Haswell CPU from Intel. Read more…

By Tiffany Trader

Leading Solution Providers

Nvidia Sees Bright Future for AI Supercomputing

November 23, 2016

Graphics chipmaker Nvidia made a strong showing at SC16 in Salt Lake City last week. Read more…

By Tiffany Trader

BioTeam’s Berman Charts 2017 HPC Trends in Life Sciences

January 4, 2017

Twenty years ago high performance computing was nearly absent from life sciences. Today it’s used throughout life sciences and biomedical research. Genomics and the data deluge from modern lab instruments are the main drivers, but so is the longer-term desire to perform predictive simulation in support of Precision Medicine (PM). There’s even a specialized life sciences supercomputer, ‘Anton’ from D.E. Shaw Research, and the Pittsburgh Supercomputing Center is standing up its second Anton 2 and actively soliciting project proposals. There’s a lot going on. Read more…

By John Russell

Tokyo Tech’s TSUBAME3.0 Will Be First HPE-SGI Super

February 16, 2017

In a press event Friday afternoon local time in Japan, Tokyo Institute of Technology (Tokyo Tech) announced its plans for the TSUBAME3.0 supercomputer, which will be Japan’s “fastest AI supercomputer,” Read more…

By Tiffany Trader

IDG to Be Bought by Chinese Investors; IDC to Spin Out HPC Group

January 19, 2017

US-based publishing and investment firm International Data Group, Inc. (IDG) will be acquired by a pair of Chinese investors, China Oceanwide Holdings Group Co., Ltd. Read more…

By Tiffany Trader

Dell Knights Landing Machine Sets New STAC Records

November 2, 2016

The Securities Technology Analysis Center, commonly known as STAC, has released a new report characterizing the performance of the Knight Landing-based Dell PowerEdge C6320p server on the STAC-A2 benchmarking suite, widely used by the financial services industry to test and evaluate computing platforms. The Dell machine has set new records for both the baseline Greeks benchmark and the large Greeks benchmark. Read more…

By Tiffany Trader

What Knights Landing Is Not

June 18, 2016

As we get ready to launch the newest member of the Intel Xeon Phi family, code named Knights Landing, it is natural that there be some questions and potentially some confusion. Read more…

By James Reinders, Intel

Is Liquid Cooling Ready to Go Mainstream?

February 13, 2017

Lost in the frenzy of SC16 was a substantial rise in the number of vendors showing server oriented liquid cooling technologies. Three decades ago liquid cooling was pretty much the exclusive realm of the Cray-2 and IBM mainframe class products. That’s changing. We are now seeing an emergence of x86 class server products with exotic plumbing technology ranging from Direct-to-Chip to servers and storage completely immersed in a dielectric fluid. Read more…

By Steve Campbell

KNUPATH Hermosa-based Commercial Boards Expected in Q1 2017

December 15, 2016

Last June tech start-up KnuEdge emerged from stealth mode to begin spreading the word about its new processor and fabric technology that’s been roughly a decade in the making. Read more…

By John Russell

  • arrow
  • Click Here for More Headlines
  • arrow
Share This