Entries filed under “ISC’10 Feature Stories”

Original features from insideHPC editors and contributors about the people, organizations, and technologies fueling the buzz around ISC’10.

The Rich Report: InfiniBand Rocks the TOP500

On my recent trip to ISC10 in Hamburg this month, the real buzz centered around the industry reaching sustained Exaflops in the next decade or so. And while this milestone is daunting enough in terms of processing, it struck me that the challenges of getting that many cores connected with low latency is going to be the highest hurdle of all. Is InfiniBand going to get us there, or will it require something way beyond current technology trends? To find out, I caught up with Brian Sparks and Clem Cole from the InfiniBand Trade Association, to talk about IBTA’s latest performance roadmap.

insideHPC: This is TOP500 week. How did IB fare in the rankings?

briansparks

Brian Sparks, IBTA Marketing Working Group Co-Chair

Brian Sparks: We did really well. InfiniBand is now deployed on 208 systems of the TOP500 sites, and that’s an increase of 37 percent from a year ago. And that success is really reflected in the upper tiers. So in the Top100, we have 64 systems. And in the Top10, we have five systems. InfiniBand has powered Top10 systems on every list since 2003, so what you’re seeing is our momentum continuing to increase.

insideHPC: Are you gaining TOP500 “share” at the expense of proprietary interconnects?

Brian Sparks: A lot of what it’s eaten has been the proprietary interconnects. If you look at all the proprietary links combined, I think it’s only 25 to 28 clusters on the TOP500. The remaining gains have come from GigE, which has gone down to the 235-240 range. There’s also a couple of 10 GigE clusters entering mix now finally.

Clem Cole, Senior Principal Engineer, Intel

Clem Cole, Senior Principal Engineer, Intel

Clem Cole: We aren’t here to bash anyone, but I think what Brian describes is correct. I’m enough of a gray-haired historian on this subject to say that the proprietary guys face a shrinking value proposition. That doesn’t mean that there isn’t room for them. I absolutely believe that for the Top5 kind of customers, there will tend to be a value for somebody to invent something that takes it to the next level. Will they do it by starting with IB, or will they start by blowing up IB like IB did with Ethernet? I don’t know. If they do go on their own, history has shown that it’s very tough to survive. The point is, I think that for the bulk of the Top500, the guys who really want a multicomputer site environment with a multi-vendor ecosystem, that’s what IB is about.

insideHPC: What are the highlights of the latest IB roadmap?

Brian Sparks: The bottom line is really that IB continues to evolve with leading bandwidth performance and other enhancements. IBTA’s current roadmap calls for 4x EDR ports at 104Gb/s data rates in 2011. That’s 26Gb/s per lane and that’s a significant uptick from IBTA’s previous roadmap from June 2008, where we projected 4x EDR at less than 80Gb/s data rate in 2011. The other performance news is that the spec is now moving from 8b/10b to a 64/66 encoding so the data throughput will now better match the link speed. So taking the new encoding into effect, EDR will be over 3X the data bandwidth that QDR now provides.

Clem Cole: I think the big message is that the performance gains are going to keep coming. So if you look at the gains we’re making from current speeds of 40Gb/s to over 100Gb/s in 2011, IB continues to be a very economical solution to one of the toughest problems in High Performance Computing.

insideHPC: Clem, you were part of the IBTA effort from the beginning. Can you tell me more about where IB came from?

Clem Cole: Well, you really need to go back to the beginning. What we think of as clustering today is really an old idea that goes back to Gerald Popek at UCLA in the early 70’s. Jerry was the first one to put his finger on this idea of taking multiple computers and orchestrating them a one big system.

So we could all see that this was the way the industry was going. And along the way, people started building these custom interconnects and we had a situation where the big vendors all had their own proprietary network technology going. At DEC we had something called Memory Channel and we were facing big development costs, like $60-$100 Million for the next generation.

Remember that this is all driven by money. So we as vendors all needed a high-bandwidth technology that was going to meet the demand, but not a one of us was getting make any money if we continued to fight each other. So InfiniBand originated from the 1999 merger of two competing designs: Future I/O, developed by Compaq, IBM, and HP, and Next Generation I/O, developed by Intel, Microsoft, and Sun.

So in the end what made InfiniBand, and for that matter Ethernet go, was everybody agreeing that this was good enough. So instead of trying to differentiate with the interconnect, we could do our own value-add in other areas.

insideHPC: How important has the Open Fabric project been to industry adoption of InfiniBand?

Clem Cole: I think OpenFabrics was one of the most important milestones in getting IB to be successful. The OpenFabrics Alliance took the OpenFabrics Enterprise Distribution upstream and helped make it part of the Linux kernel. So now you’ve got the distributions like SuSe and Red Hat where it’s just in there. And don’t forget Windows as well.

insideHPC: InfiniBand has a reputation of being much harder to implement and manage than Ethernet. Does the IBTA recognize this as an issue that needs addressing?

Brian Sparks: I think that for HPC, IB has gone through a lot of evolution in terms of ease of use. When it first came out, you had a lot of scientists who were eager to play around with things and make it work. And now as you start going into enterprise solutions, people just want to drop it in and not worry about it. So as the years have evolved, we’ve been able to make that possible.

As an organization, IBTA has been trying to address InfiniBand’s reputation as being difficult to work with. We recently came out with a eBook called Introduction to InfiniBand for End Users. It’s kind of an IB for Dummies document with some key know-how such as what does IB management look like and how does that differ from what you’re used to in terms of Ethernet management.

So InfiniBand continues to evolve, and these efforts are really important because IB isn’t just for supercomputers and hard-core scientists any more. IB lets you add a server any time you want, and for things like cloud computing that’s a great value to the enterprise as well.

Also posted in Events, HPC, HPC Hardware, ISC'10 Feature Stories, ISC10, Network | Leave a comment

Video: Supermicro Showcases Twin Server Line

Coming from a hardware background, I love to see the latest gear at ISC. Chips, sockets, heat sinks, and blades: this is the kind of stuff I geek out on. So last week I dropped in on the Supermicro booth with my camera because they always fill entire walls with mother boards from their latest server products.

In this video, Supermicro VP of Marketing Don Clegg gives us a tour of their new Twin family of servers.

Also posted in Compute, Events, Featured Stories, GPUs, HPC, HPC Hardware, ISC'10 Feature Stories, ISC10, Video | Leave a comment

Between the ISCs: looking back over the past 12 months of HPC accomplishment

Contribution by regular readers Thomas Sterling (an HPC Rock Star) and Chirag Dekate of Louisiana State University. This article follows Sterling’s review of the past 12 months in HPC, given each year at ISC in Germany.

As the field of HPC enters its second decade of the 21st Century, new directions in system structure, operation, and programming are being driven by the technical trends and application needs at extreme scale.

Thomas Sterling

Unlike never before, even with the expectation of the continuance of Moore’s law, the opportunities of performance gains are threatened by the second turning of the decades’ long S-curve HPC has been traversing. This last year has seen dramatic evidence of the initial flattening with the imposition of power and complexity constraints as well as innovative approaches and market products to address them. At ISC 2010 in Hamburg, Germany, the authors were afforded the opportunity to review the events that best reflect the trends, directions, and accomplishments of the last year by the international supercomputing community: industry, academia, and national facilities and programs. The chosen theme highlighted for this year’s presentation on the state of the field in HPC was “Igniting Exaflops” to underscore and acknowledge the major steps that have been taken over the intervening 12 months to prepare the international community for a future of Exascale computing before the end of this decade. But first, let’s summarize some of the recent key achievements in HPC and their impact as taken from the 7th annual ISC retrospective.

In brief, hex cores for multicore in 32.0 nanometer fabrication technology have become mainstream replacing last generation quad core chips for new product offerings based on commodity clusters that continue to gain market share with respect to MPPs. Sockets combing multiple dies are becoming available with up to 12 cores in cache coherence structure SMPs. Heterogeneous system structures are gaining traction with the increased integration of GPUs for floating point intensive applications.

Equally important in this direction are the advances in programming methodologies, with improved system software merging conventional APIs and CUDA or OpenCL making this emergent class of HPC systems of greater utility to the technical computing end users.

A major competition has been waged in the field of networking for clusters between Ethernet and Infiniband, with Ethernet representing the larger deployed base, but Infiniband dominating the high end systems as well as the total aggregate performance across the Top-500 list. Many applications of scientific and technical importance have been developed, pushing new discovery forward with the first significant Petaflops scale applications running on such machines as Jaguar at Oak Ridge National Laboratory and recognized by the Gordon Bell Prize. Green computing has continued to gain attention with advanced designs and techniques being applied to reduce overall energy requirements and limit the upward surge of peak power demand.

The Top500 and the race to the top

Jaguar at Oak Ridge National Laboratory is still the fastest supercomputer in the world as measured by the Linpack Benchmark (some systems are not similarly rated or reported). With a sustained performance of 1.76 delivered Petaflops (and higher on some applications), this integration of Cray XT4 and XT5 subsystems based on AMD Opterons runs SUSE Linux operating system, an array of compilers from multiple software vendors, and offer support for diverse programming models.

But a new contender for second place comes from ShenZhen, China, and with a Linpack performance of 1.27 Petaflops it handily exceeds the coveted 1 Petaflops threshold. This system uses a heterogeneous system architecture of Dawning TC3600 blades with IntelX5650 processors, and Nvidia Tesla C2050 (Fermi) GPUs. Indeed, the system’s peak performance of nearly 3 Petaflops exceeds that of Jaguar itself.

Roadrunner at LANL, the first Petaflops computer, is now entering its third year of operation using a heterogeneous architecture that incorporates IBM Cell processors with conventional AMD Opterons. Also at Oak Ridge is another Cray system, “Kraken”, that just breaks a Petaflops peak capability using dual hex core Opterons and the advanced Cray SeaStar2+ router. Germany’s Jugene IBM BG/P st Julich also exhibits Petaflops peak performance with almost 300,000 PowerPC 450 cores. China retains its Tianhe system that also peaks above a Petaflops with a cluster combing Intel Xeon and AMD GPUs. Other systems worth note are Russia’s Lomonosov and Shaheen in Saudia Arabia, with both providing hundreds of Teraflops. It should be noted that this year it was Hewlett-Packard that has deployed the largest number of HPC systems, beating out IBM for the top slot. No other supplier even comes close in this market to these two giants.

The year in cores

The foundation of all of each of these super systems is their processor cores, and this year has seen significant advances from the semiconductor component manufacturers.

Intel dominates HPC system deployment and total aggregate performance with a number of slightly different offerings. The Westmere 2-core and 6-core X5600 processors are implemented in 32 nanometer technology. The IBM Power7 architecture is in 45 nanometer, with one of the largest processor dies ever, and pushes clock speed to above 4 GHz. This 8-core package will deliver a maximum of 265 Gigaflops and incorporates advanced pre-fetching of data and instructions. It will be integrated in the Blue Waters machine to be delivered to UIUC next year. The 8- and 12-core AMD Magny-Cours processor (in 45 nanometer technology) uses HyperTransport 4 inter-core communication technology for more efficient cache coherence.

But what of Itanium? In the keynote address by Intel representatives at ISC 2010, no mention was made of its role, although it is known that a future roadmap exists with targets of Poulson in 2012 and Kittson in 2014. However, this year both Microsoft and Red Hat have announced that they will stop supporting this architecture. HP, one of the originators of much of the Itanium design, is expected to continue to deliver products based on the platform.

Also of note: Rock, Sun’s next-generation processor architecture, was terminated during the last year.

Accelerators

Nvidia has delivered its new GPU, Fermi, for improved double precision performance, and is making major strides in releasing improved CUDA and OpenCL software for programmer support. AMD has also advanced its ATI accelerator with the release of Cypress (RV 870) with better than half a Teraflops double precision peak performance.

HPC people

Individual achievements are acknowledged. Ken Miura of Fujitsu was given the Cray Award for his work in vector computing. The Fernbach Award was to Roberto Car and Michele Parrinello for their joint method in molecular dynamics. And the inaugural Kennedy Award was presented to Francine Berman of Rensselaer Polytechnic Institute for her pioneering work in building a national grid based cyberinfrastructure in the US. William Gropp of UIUC was awarded this year’s IEEE TCSC Medal for Excellence in Scalable Computing. He was also recently elected to the US National Academy of Engineering.

With sadness we also note the passing of John Mucci, formerly of Digital Equipment Corporation and a cofounder of SiCortex in 2002.

Getting to exascale

This year also saw the inauguration of the first sponsored programs in Exascale computing.

The International Exascale Software Project (IESP) has involved participants from North America, Europe, and Asia to establish a world-wide coordinated activity to develop the software infrastructure needed in preparation for Exaflops computer architectures targeted for deployment by the end of this decade. The IESP held major technical congresses were held over the last year in France, Japan, and the UK to develop a joint international roadmap.

It is recognized by many (there is controversy on this point) that methods and means for realizing Exaflops scale computing will out of necessity prove very different from those which have successfully brought the field in to the Petaflops era. It has been well understood that historically software has always lagged behind hardware, but this time software must precede hardware both so that we will be ready to use such systems when they are developed, and to inform that development through understanding of software needs.

A second initiative that has been undertaken that will lead to technologies that can be applied to Exascale system deployment is the US DARPA UHPC (Ubiquitous High Performance Computing) program. Although not explicitly established for this purpose, UHPC will produce prototypes of Petaflops racks within the power budget of 60 Kilowatts that could be integrated into full Exascale systems by the end of this decade.

Proposals have been submitted, and DARPA should be announcing the winners before next month. This is a very exciting program with a very real prospect of reinventing how future scalable computing will be achieved.

The US DOE has also launched some new programs relevant to Exascale computing, including one to realize the goal of an X-Stack, the software infrastructure that will be required for Exaflops computing. This program has already received proposals, and will be announcing selected investigators in the near future.

Together, these and other programs begun this year, along with many technical workshops that have also been conducted within the last twelve months, are rapidly putting the world on track to aggressively and effectively move all aspects of system development forward towards the performance goals of the year 2020.

This year has been one of significant product advances, application accomplishments, and initiation of important pathfinding work. The coming year is anticipated to be even more valuable.

Thomas Sterling

Dr. Thomas Sterling is a Professor of Computer Science at Louisiana State University, a Faculty Associate at California Institute of Technology, a CSRI Fellow for Sandia National Laboratories, and a Distinguished Visiting Scientist at Oak Ridge National Laboratory. He has also been recognized as an HPC Rock Star by insideHPC.

Chirag Dekate

Chirag Dekate is pursuing a PhD at LSU; his topic is resource management and scheduling of dynamic data driven graph executions.

Also posted in Events, Featured Stories, ISC'10 Feature Stories, ISC10 | 1 Comment

Video: Life is Random, So is Your Storage

I am fascinated with the CGI effects in today’s feature films and it seems like HPC infrastructure has become a competitive weapon for the movie studios. So when I heard that Weta Digital (Lord of the Rings) was using BlueArc, I decided to take a closer look.

In this video, BlueArc’s Director of HPC Marketing Bjorn Andersson talks about why storage loads are so random and how the company’s storage solutions are built for optimal performance.

Also posted in Events, Featured Stories, HPC, HPC Hardware, ISC'10 Feature Stories, ISC10, Storage, Video | Leave a comment

Video: Why is Everyone Talking About GPUs for HPC?

I think that NVIDIA was a big winner this year at ISC, and it wasn’t just because China’s new Tesla-powered Nebulae Supercomputer came in at number 2 on the TOP500. The reason for me was rapid adoption; I visited at least half a dozen booths that featured the latest NVIDIA GPUs in a variety of configurations. Clearly, the company is getting traction in the market.

In this video, I interview Andy Keane, NVIDIA General Manager of the Tesla Business Unit and discuss the advantages of GPUs for HPC. He also gives us his views on power efficiency and Exascale computing.

Also posted in Events, Featured Stories, GPUs, HPC, HPC Hardware, ISC'10 Feature Stories, ISC10, Video | 2 Comments

Video: New Modeling and Simulation Leadership Panel Seeks Members

In this video, Addison Snell, CEO of Intersect360 Research, describes the formation of the Modeling and Simulation Leadership Panel, a “worldwide panel of organizations using computational modeling, simulation, and analytics to advance their cutting-edge positions in engineering development and research.”

It looks like a sweet deal to me. End users who join the panel will receive free access to research from Intersect360 Research. In return, they agree to do a 30 minute survey on a quarterly basis.

For more information, check out the Modeling and Simulation Leadership Panel site.

Also posted in Events, Featured Stories, HPC, ISC'10 Feature Stories, ISC10, Video | Leave a comment

Video: Dynamic GPU Reassignment with NextIO vCORE Appliance

I had grand designs to tape a bunch of demos at ISC10, but time just wouldn’t allow. So after scouting around a bit, I decided to film the best demo that I could find.

In this video, Kyle Geisler from NextIO demonstrates the company’s vCORE Appliance, the “world’s first flexible platform for GPU reassignment in the HPC datacenter.” Watch as he moves GPU resources around on-the-fly, even as they continue to run applications. Consider me impressed, and I can just imagine how powerful this capability will be for putting GPUs to work in the cloud.

Also posted in Datacenter operations, Events, Featured Stories, GPUs, HPC, HPC Hardware, ISC'10 Feature Stories, ISC10, System Management, Video | Leave a comment

Video: For HPC, The Russians are Coming!

There was something very new this year at ISC. At the very front of the exhibit hall was a large booth from T-Platforms, a Russian HPC vendor that drew a lot of attention with their hospitality and innovative approach to high performance computing. They’re young, they’re enthusiastic, and HPC is all they do.

In this video, T-Platforms VP of Marketing Alexey Nechuyatov describes the company’s focus on HPC and how they plan to bring their solutions to market in Western Europe.

Also posted in Events, Featured Stories, HPC, ISC'10 Feature Stories, ISC10, Video | Leave a comment

Video: My iPad vs. IBM Blue Waters

Thanks to John West, I was able to embed this video from the ISC video blog coverage. I was watching the video and there I was on my iPad, which is kind of an odd feeling. Just so I don’t have to settle for that one second of fame, a couple of people told me that my iPad was the coolest new technology they saw at the conference.

In the beginning montage, you’ll see a brief shot of me working with my iPad. Then Heike Jagode, a computer scientist at the University of Tennessee, talks with Jack Dongarra about exascale software. After that, Stuttgart Supercomputing Chief Michael Resch gives his views on the conference. Later on, from Professor Thom Dunning from NCSA shows off the prototype “IH Server Node” from their 10 Petaflops Blue Waters project.

Heike Jagode and John Shalf, head of NERSC’s Advanced Technology Group, did a great job of covering the show. The rest of their videocasts are now posted for viewing at the ISC conference site.

Also posted in Events, Featured Stories, HPC, ISC'10 Feature Stories, ISC10, Video | 1 Comment

The Rich Report: Five Minutes with Cray’s CTO, Steve Scott

Now that ISC has turned 25 and I’m seeing all these familiar faces, I’ve been getting a little nostalgic for the Old Cray Days. And as I look around the show floor at all this capitalism going on, I think about how it all goes back to Seymour Cray, the guy who created the supercomputer industry. So this week it was a great pleasure for me to meet with Cray CTO, Steve Scott.

insideHPC: You know, I started my career at Cray back in 1986, so I just wanted to say how wonderful it is to see the company doing so well and making money again.

Steve Scott

Steve Scott

Steve Scott: It is nice. A lot of the credit I think goes to our CEO, Peter Ungaro. He knows the industry inside and out and he put together a management team that got the finances in order. That was a big deal because people would go, “We like what you’re saying, but what about the financial viability of the company?” So we don’t get that any more.

insideHPC: So I’d like to start out by talking about Exascale. I was having a discussion earlier and someone asked which company is going to get there first. It seems to me it’s down to Cray or IBM.

Steve Scott: You certainly can’t count them out. But there’s actually a record to consider. If you go back and look at not the first peak number or linpack result, but the first sustained application Gigaflop, Teraflop, and Petaflop, they were all on Crays. So the first sustained application Gigaflop was on a CRAY Y-MP in 1988. The first Teraflop was in 98 on a CRAY T3E. And the first sustained application Petaflop was in 2008 on Jaguar and the CRAY XT-5. So I’ve gone out on a limb and said publically that the first sustained application Exaflop will be on a Cray in 2018. So that’s our internal target. And this one is going to be harder than the last one. So now that Cascade is kind of in the bag, most of my time is spent thinking about how we get to Exascale.

insideHPC: So what is the superscale user community doing right now to get things going?

Steve Scott: There’s a lot of stuff heating up in the Exaflop race or whatever you want to call it. The DOE Exascale program is getting closer to reality and the DARPA Ubiquitous High Performance Computing program just got off the ground. They’re not specifically targeting Exascale, but UHPC actually grew out of all the study teams that DARPA sponsored: one on hardware, one on software, and one on resilience.

So UHPC is focused on a Petascale in a box, but at power efficiency targets that are good enough for an Exascale. They target 50 sustained Gigaflops per Watt and if you scale that up, that turns into Exascale for 20 Megawatts. So if we can hit the UHPC target, we will be on our way.

Interestingly, the biggest datacenters today are north of 100 Megawatts. That’s the big Internet datacenters, you know Google, Amazon, and Microsoft. So this Exascale system would be 20 Megawatts, which is bigger than any single system today. Jaguar (#1 on TOP500) is the biggest HPC power consumer today at roughly 7 Megawatts, but it’s also the greenest x86 system in the TOP50 in terms of sustained flops per watt.

insideHPC: So if they can hit these targets, the Exascale system would consume roughly three times as much power as Jaguar?

Steve Scott: Yes, and that’s a big factor because you have to consider the budget to run the system and what’s practical from a political perspective. The power budget for a 20 Megawatt system would be roughly $20 Million per year, since power costs about a million dollars per Megawatt year.

insideHPC: It sounds like the next 8 years are going to be very exciting.

Steve Scott: It is fun. And the past five or so years have been pretty interesting as well. We had this huge technology inflection point, you know, with power and delay and transistor counts are going up, a bunch of things came together which said we’ve got to do things completely differently. That’s why we had the whole multicore phenomenon happening.

But when we look out to Exascale, it’s clear that just straight multi-core x86 is not going to get us there. Fundamentally we have to do something different underneath the covers to get the energy efficiency. So that’s why we’re pretty interested these days in looking at forms of accelerated computing. We formed a partnership with NVIDIA and we’re working with them on some future stuff. It’s not going to take over the world overnight, but it’s headed in the right direction.

insideHPC: So what are the most important things you would like potential customers to know about Cray?

Steve Scott: I would say number one is Efficiency. We’re all about power efficiency and system efficiency. Our focus is entirely on sustained computing, sustained results for real scientists. So we work very, very closely with our end users. We have a relationship with them that is not typical. We know their codes and we work on things that don’t make the peak of the machine any better, but things that help them get sustained results. So the fact that the Jaguar system is sitting at number one on the list is nice, but we really don’t care how fast it runs linpack. There are three applications on Jaguar that are achieving over a Petaflop, sustained, and that’s what it’s all about.

I would say the third thing is that Cray now has complete coverage of the HPC space, from the desk-side up to the biggest supercomputers. We’ve got a collaboration with Microsoft where we offer Windows HPC on our low-end, desk-side boxes. And we have now introduced the CX1000, which is a single rack system that uses Intel processors. It has SMP nodes as well as distributed-memory nodes. It also has graphic acceleration, all integrated and ready to run. So HPC is all we do and we can get you started.

Also posted in Events, Featured Stories, HPC, ISC'10 Feature Stories, ISC10 | 3 Comments

Exclusive: Russia’s biggest HPC provider talks Rusnano and the bid to spur innovation with HPC

insideHPC talks with Vsevolod (Seva) Opanasenko, CEO of T-Platforms Group

T-Platforms logoT-Platforms Group, headquartered in Moscow, has quietly been expanding its installed base throughout Russia and CIS, now approaching 200 supercomputer installations, and starting to build a presence in Western Europe. The company’s next-generation operating system, Clustrx, combined with an innovative architectural approach, is getting significant attention in the HPC community.

Last week they announced the award of a $6 million nanotechnology and supercomputing enablement program funded by the Russian Corporation of Nanotechnologies, Rusnano. The program is somewhat unique and could stand out as a model for other countries in that it is designed to create real-world adoption of advanced HPC and nanotechnologies in production environments.

Neither T-Platforms nor Rusnano are very well known throughout this community, so we wanted to give our readers more insight to both this announcement and the organizations behind it. We were fortunate to catch up with Vsevolod (Seva) Opanasenko, CEO of T-Platforms Group, and as a result, are pleased to bring you this feature interview.


insideHPC: Can you tell us a little more about Rusnano?

Vsevolod (Seva) Opanasenko: Rusnano is the Russian Corporation of Nanotechnology and was established by the Russian government in 2007 as a state corporation chartered to co-invest in nanotechnology and supercomputing industry projects that have high commercial potential or social benefit. The corporation also provides scientific and educational programs to help ensure the success of the various investment projects.

Collaborative funding from Rusnano is very specific toward the development of actual real-life simulations to bring companies immediate value and benefits — accelerating the evaluation and analysis of the impact of supercomputing to industrial and scientific discovery. Overall, the Rusnano program goals have been established to eliminate technical barriers to adoption and to improve production and productivity.

insideHPC: What are some of the specific deliverables from this program, and what level of support will organizations receive?

Seva: Well first, T-Services will oversee the project management for this program. We will accept and review the many submissions to determine their feasibility based on a number of criteria such as potential importance, practicality, possible ROI, and applicability to spawning commercially sound results or products.

There are two engagement scenarios we anticipate as being ideally suited for this program: organizations that have appropriate computational tasks identified but have no expertise or computational capacity, or those who have the expertise in software and need only the computational capacity.

We will provide extensive consulting expertise from helping to fully define the computational problems being addressed, to preparing the task budgets, and finally selecting the most promising tasks to be presented to the Rusnano Expert Council. Then, once an organization is engaged in the program, T-Services will provide, as necessary, the actual modeling, simulation and analysis of the computational tasks. This in itself makes this program quite unique — setting it apart from other programs in which organizations are granted use of computers — but nothing at this level of deep technical support.

It’s actually quite amazing to think that the program will fund the efforts for 40 different tasks — half of the selected tasks will be directly involved with nanotechnology, or nanoscience, and half will be in support of engineering efforts in areas such as shipbuilding, aerospace, oil and gas, chemistry, pharmaceuticals, energy and construction.

insideHPC: Could you expand on that just a bit. What kind of tasks?

Seva: Well, think of this as a nanotechnology / HPC industry watershed. Essentially, half of the tasks should come from industrial companies and organizations engaged in actual, real-life production or manufacturing, while the other 50% can come from the organizations involved in science and R&D. The bottom line though is that these task — or projects — should be aimed at research and development of technologies that can subsequently be commercialized. So, for example, an organization might have a very creative approach requiring a new computational process that could greatly assist in the search for new oil or gas deposits. But, the type or capability of computer power may be way out of their reach, or they may not even have the in-house expertise to develop the software for modeling and simulating the process. If the potential for this task looks promising and fits the program criteria, they could in theory receive funding that would provide both the computational cycles and the deep technical expertise necessary to proceed to a proof of concept stage — and even on to production.

insideHPC: Will this program have an impact on HPC adoption outside of Russia?

Seva: HPC adoption, in general, and having nothing to do with geography, faces many barriers when it comes to commercial organizations. HPC systems, historically, have not been practical for organizations without specific expertise in this area of computing. The lack of talent such as the skilled resources needed to develop modeling and simulation software is a deal breaker for many companies. The high Total Cost of Ownership (TCO) often associated with HPC systems has made it extremely difficult for smaller and mid-size companies to justify moving in this direction — especially without a proof of concept, which is essentially impossible to provide in the absence of professional end-to-end technical services.

Programs such as the U.S. Department of Energy’s INCITE program, will give away approximately 1.6 billion supercomputer processor hours in 2011. This is an amazing commitment on the part of the U.S. Federal Government to accelerate scientific discovery. However, and not to downplay the vital importance of this program in any way, but this type of program only helps those organizations who already have extensive modeling and simulations expertise in place. In contrast, the Rusnano program is designed to engage companies with little or no expertise in these areas of advanced computing. An organization only has to have an idea of how relatively large-scale computation, modeling and simulation could help. The organization doesn’t have to boast expertise in software or even have CAD models. In Russia, this will help to ignite many innovative tasks with companies who would not get there on their own. But, beyond Russia, this is an example of market building on a much broader scale. The Rusnano program will bring HPC into many new companies to advance discovery in nanotechnology, a number of scientific research disciplines, and critical production areas requiring advanced computation — with both the supercomputing physical resources and the technical expertise necessary to make it happen.

insideHPC: So, the contract described in this announcement is between Rusnano and T-Services, which I understand is a T-Platforms Group company. Can you tell us a little more about T-Services?

Seva: You are correct in that T-Services is one of the T-Platforms Group companies. The easiest way to describe the unique value proposition of T-Services is to discuss it in the context of T-Platforms.

T-Platforms provides solution integration consulting to assist with basic HPC enablement such as hardware integration and system configuration as well as software optimization focused on helping customers improve application performance. The technical benchmarking and integration specialists at T-Platforms work with customers to develop optimum hardware and software configurations, matching the unique requirements of the user, along with application software optimization to achieve improved application performance where necessary.

T-Services however provides end-to-end simulations and modeling for customers using a wide range of traditional HPC applications. The company’s deep technical expertise includes disciplines such as CFD and structural analysis. T-Services also offers supercomputer center management services to deliver maximum effective use of all HPC systems to improve an organization’s ROI, as well as on-demand computing services from simply allocating compute cycles to implementation, operations and administration of commercial software packages.

The powerful combination of T-Platforms’ flexible hardware architecture and technical system management expertise, combined with the end-to-end modeling, simulation and technical services available from T-Services, enables organizations to achieve time-to-solution advantages while avoiding pre-packaged, pre-configured systems ill-suited to their unique challenges.

T-Services delivers value to organizations who only need HPC occasionally — or would rather outsource certain parts of the HPC simulation workflow.
Another level of support offered by T-Services is management of supercomputer sites. T-Services can act as a management company, taking responsibility for all resources and providing not only the operation and administration talent, but also managing the workflow with customers, from marketing and sales to actually doing the technical tasks.

insideHPC: And can you elaborate just a bit on the working process of this program and the role T-Services will play in the overall program management?

Seva: Sure. This program is actually very well structured and organized. The first stage is the initial evaluation and selection phase. T-Services will interface with all the applicants to help define the proposed tasks, identify and choose appropriate software, establish the proposed working budgets for each task, and generally organize the selected applicants and task descriptions, along with all the requirements for submission to the Rusnano Expert Council.

Then, for the next stage, members of the Rusnano Expert Council will review the submissions based on the defined criteria which I can break down into six pieces.

  • Does the project fit the scope of industries authorized for this program?
  • Is HPC really needed and applicable?
  • Is the project innovative — and could it lead to creation of an original technology?
  • Can this project be co-financed by the author of the problem?
  • Specific to industry — how would any potential profit from solving this computational problem be measured for the organization?
  • Specific to research organizations — might this research lead to creating a technology with a good commercialization potential?

insideHPC: What organizations are eligible for consideration within this program and how would they apply? What do they need to know?

Seva: This program is being launched within Russia only. Any organization, either research or industrial, nano or other qualified industries, may apply, but their ideas should meet the criteria of Rusnano that we’ve mentioned. Organizations should note that the Rusnano Expert Council includes representatives from Rusnano, experts in modeling in engineering and nano applications (not ISVs but users), experts from industry and nanoscience (to evaluate the potential benefits to industry and the degree of innovation of the projects), a representative from the government, and even a representative of a T-Services competitor, to control budget evaluations and methods chosen by T-Services. The organizations apply first to T-Services through the web-site, submitting their ideas and explaining their correspondence to the criteria, and then after the evaluation process mentioned earlier, T-Services submits the chosen projects to the Rusnano Expert council.

insideHPC: How long will this program run? What will happen when the program ends?

Seva: Overall, 10 months. The expert council gathers several times through the project but it should select all the 40 tasks within 5 months from the project start — otherwise some simulation tasks that take a couple of months to complete might not be fulfilled in the program’s identified window of ten months. The preliminary results should be presented and evaluated seven months from the project start.

So, there should be at least 40 computational problems corresponding to criteria and approved by the council. The level of co-financing from the participating organizations will be an important measure of success. And finally, upon completion, Rusnano will consider whether it is worthwhile to build a supercomputer center for collective use, for nano and other industries.

Also posted in Events, Featured Stories, ISC'10 Feature Stories, ISC10, National and Legislative Action | 1 Comment

Top500 Movers and Shakers

For those of you interested in lists, the June 2010 Top500 list was released this morning at ISC.  There was some serious movers and shakers with respect to the top ten participants.  Most obvious was China’s move to number two on the list.  The Nebulae system was built by Dawning with a combination of Intel Nehalem 5650 cores and NVIDIA Tesla C2050 GPUs.  C2050′s?  Yes, they’re Fermi’s.

China’s ambition to enter the supercomputing arena have become obvious with a system called Nebulae, build from a Dawning TC3600 Blade system with Intel X5650 processors and NVidia Tesla C2050 GPUs. Nebulae is currently the fastest system worldwide in theoretical peak performance at 2.98 PFlop/s. With a Linpack performance of 1.271 PFlop/s it holds the No. 2 spot on the 35th edition of the closely watched TOP500 list of supercomputers.

Roadrunner at LANL dropped to number three on the list.  It holds at 1.04PF.  This pushed Kraken at ORNL/NICS to number four.  The number five system on the list was the powerful system in Europe.

At No. 5 is the most powerful system in Europe — an IBM BlueGene/P supercomputer located at the Forschungszentrum Juelich (FZJ) in Germany. It achieved 825.5 teraflop/s on the Linpack benchmark.

A few more highlights from the release include:

  • The entry level to the list moved up to the 24.7 teraflop/s mark on the Linpack benchmark from 20 teraflop/s six months ago. The last system on the newest list would have been listed at position 357 in the previous TOP500 just six months ago. This replacement rate was far below average. This might reflect the impact of the recession and purchase delays due to anticipation of new products with six or more core processor technologies replacing current quad-core based systems.
  • Quad-core processor based systems have saturated the TOP500 with now 425 systems using them. However, processor with six or more cores per processor can already be found in 25 systems.
  • A total of 408 systems (81.6 percent) are now using Intel processors. This is slightly up from six months ago (402 systems, 80.4 percent). Intel continues to provide the processors for the largest share of TOP500 systems. The AMD Opteron is the second most common used processor family with 47 systems (9.4 percent), up from 42. They are followed by the IBM Power processors with 42 systems (8.4 percent), down from 52.
  • IBM and Hewlett-Packard continue to sell the bulk of systems at all performance levels of the TOP500. HP lost its narrow lead in systems to IBM and has now 185 systems (37 percent) compared to IBM with 198 systems (39.8 percent). HP had 210 systems (42 percent) six months ago, compared to IBM with 186 systems (37.2 percent). In the system category, Cray, SGI, and Dell follow with 4.2 percent, 3.4 percent and 3.4 percent respectively.
  • IBM remains the clear leader in the TOP500 list in performance with 33.6 percent of installed total performance (down from 35.1 percent), compared to HP with 20.4 percent (down from 23 percent). In the performance category, the manufacturers with more than 5 percent are: Cray (14.8 percent of performance) and SGI (6.6 percent), each of which benefits from large systems in the TOP10.
  • The U.S. is clearly the leading consumer of HPC systems with 282 of the 500 systems (up from 277). The European share (144 systems – down from 152) is still substantially larger then the Asian share (57 systems – up from 51). In Europe, UK remains the No. 1 with 38 systems (45 six months ago). France passed Germany and has now 29 (up from 26). Germany is still now the No. 3 spot with 24 systems (27 six months ago). Dominant countries in Asia are China with 24 systems (up from 21), Japan with 18 systems (up from 16), and India with 5 systems (up from 3).

For more info on the movers and shakers on the Top500, read their full release here.

Also posted in Events, ISC'10 Feature Stories, ISC10 | 3 Comments

VCollab Launches Social Network for 3D Visualization

Social Networks and Cloud Computing are probably the two hottest trends in IT today. And while we like to think that HPC is all about the leading edge, it seems to me that the High Performance Computing is missing out on the party. That may have changed today with the launch of Marechi.com, the first social network for 3D Scientific Visualization.

I caught up with Prasad Mandava, CEO of VCollab, to talk about Marechi and how the company’s 3D collaboration software enables global teams to work together on complex CAE models and virtual prototypes.

insideHPC: What is Marechi.com all about?

Prasad Mandava: Today’s Social Networks are about sharing two-dimensional media. So, if you think about how sites like Flickr let you share photographs or YouTube allows you to share videos, Marechi lets you share your 3D models/simulations and socially network with like-minded engineers. This should be of particular interest to the HPC Community because now they can easily reduce, publish and share not only the static 3D data but also the interactive simulations from their scientific computations/discoveries. Marechi uses the state-of-the-art VCollab solution as the foundation for publishing 3D content and also to visualize the physics/simulations thru the free interactive VCollab 3D Viewer.

insideHPC: Wasn’t this kind of sharing always possible by just posting links to the simulation data on a web site?

Prasad Mandava

Prasad Mandava

Prasad Mandava: Certainly, but the problem with shared visualization in the CAD/CAM/CAE arena is two-fold. First, the file sizes can be very large, even up to multiple Terabytes in some cases. So often it isn’t practical to just “download the file” and view it like you would a PDF, an image, or a WORD document. That’s where Vcollab software comes in with it’s ability to shrink these models/simulations down by as much as 99 percent. So suddenly these files can be made small enough to share in an email, Powerpoint presentation, or a web site like Marechi.

Now, the second problem with shared visualization has to do with the onerous requirement to have the right equipment and software license in order to view the data. Large product development companies use hundreds of different engineering software packages to solve the physics, so that is no small task, especially in cross-discipline collaboration. Plus, some of these software packages are extremely specialized and are very difficult to use, so an exec or a downstream user like a designer or test or service engineer would have no idea how to view or manipulate the model/simulation.

VCollab solves these problems with a batch converter program that runs on remote HPC servers/clouds and turns the complex 3D models/simulations into a lightweight CAX format that users can then look at through 3D VCollab Viewers. This way, all one needs is a lightweight  VCollab viewer application on the other end. This is similar to the need of having Flash plug-in to view Flash files. VCollab also enables the MS Office and WEB integration of the Simulation data using these viewers. By publishing CAX files into Marechi, one can easily achieve the WEB integration.  Marechi provides engineers with codes to embed the published 3D simulations from Marechi into their blogs. One can easily embed the CAX files with simulations into MS PowerPoint slides or WORD documents for 3D presentations or create 3D reports.

insideHPC: So how was this accomplished before the development of VCollab technology?

Prasad Mandava: Well, consider a case where a CAE manager has to conduct a design/analysis review of a car model. After the slow, painful process of transferring the large files electronically from HPC to local systems, the CAE manager asks an analyst to capture some jpeg images and .avi videos that represent the simulation. Then the manager puts them in a PowerPoint presentation and takes it in for an analysis review. But what if somebody at the design/analysis review asked about an aspect of the model that wasn’t depicted in the images/videos?  The whole communications process gets very cumbersome. That’s a big drag on the productivity of the engineers.

insideHPC: So moving forward to today, how does a typical customer use a product like VCollab for their workflow?

Prasad Mandava: Our customers use VCollab to reduce, visualize, publish, share and communicate 3D product simulation data with their global teams and to improve 3D collaboration with down stream users as well as suppliers. For example, designers can now receive 3D simulations as portable CAX files and they can study the simulations and provide feedback to the analysts.  Suppliers / CAE service providers can share their simulation work with their customers as 3D CAX files and get the customers feedback easily.

HPC Cloud service providers and IT departments can use VCollab solutions to reduce the large simulation results files on the HPC/Cloud servers before sharing them for visualization, reducing the large file movements, saving bandwidth costs and most importantly improving the productivity of engineers. Engineers no longer need to wait for long hours to receive the data and visualize the data from the HPC/Cloud. They might also use VCollab to combine results from aerodynamics simulations and structural dynamics of an aircraft into one CAX file on HPC server before sharing.  We’ve also seen customers using the software to visualize large engine simulation results, archive automobile crash simulations, and integrate all modeling and simulation data into a 3D virtual prototype for product demonstrations and for design/analysis reviews, resulting in measurable cost savings and productivity gains.

As for analysts, they can improve their productivity and can do more work in less time with smaller files. Analysts can embed 3D simulations into PowerPoint slides and explain their work more easily to others during analysis reviews – leading to more productive design/analysis reviews and effective decision making. In a similar way, teachers can embed the 3D simulations into Powerpoint and demonstrate interactive physics to students, in the class rooms. Marechi also empowers Universities, teachers, students or even the skilled individuals to publish interactive physics / 3D simulations into Marechi and share this knowledge across the globe. This could immensely help students to understand physics/simulations and to improve the quality of science and engineering instruction. There are many such possibilities.

insideHPC: So how did the idea of 3D collaboration come together with Social Networking?

Prasad Mandava: Our company works a lot with universities and research institutions. We already have partnerships with many Universities across the globe and we intend to develop more partnerships. We wanted to provide an easy way for them to share 3D simulations/results, and today’s students are very much comfortable with the idea of Social Networking.  So the idea was to provide a virtual space where today’s brightest young minds could begin to interact with practicing engineers and work together on collaborative projects. VCollab software gets the technology out of the way and puts the right tools into their hands. We also believe that Marechi.com could be significant to the Universities that are working in the FEA, CFD and other scientific areas as they can now share their knowledge with the rest of the world to significantly impact the way students can learn science and engineering.

insideHPC: Is the Marechi.com Social Network is open to anyone?

Prasad Mandava: Yes, we are launching it as a private beta implementation as of today. Anyone can sign up for an account, so I’d like to encourage your readers to register for an account and try out some of the models that are already posted.

insideHPC: How can Universities publish data into Marechi?

Prasad Mandava: Interested universities can contact support@vcollab.com to learn how to get the publishing tools from VCollab.

insideHPC: Where can we learn more about VCollab and Marechi?

Prasad Mandava: Sharing is easy:
http://www.vcollab.com
http://www.marechi.com

Also posted in Events, HPC, ISC'10 Feature Stories, ISC10, Visualization | 6 Comments

insideHPC talks with Wolfgang Gentzsch about ISC and the world of HPC

As we prepare for the 25th anniversary of the International Supercomputing Conference (ISC’10), we once again tracked down one of our favorite commentators, Wolfgang Gentzsch, for some inside perspective.


insideHPC: ISC is celebrating its twenty-fifth anniversary. How many of the ISC conferences have you attended?

W. Gentzsch

Wolfgang Gentzsch: I’ve been fortunate to have attended all the ISC conferences since the first event. It seems hard to believe that it has been 25 years.

insideHPC: What are some of the most significant changes you have seen with the ISC conference over those years?

Wolfgang Gentzsch: Watching the growth of the conference has been just amazing. And, not only did ISC grow from about 150 attendees in 1986 to almost 2,000 today, the unique, genial atmosphere that this conference has become known for managed to survive that growth.

Looking back over the years, some of us certainly miss the unforgettable wine tours to the Mannheim hinterland in the Summer time, but of course, today that would just not be practical. Can you imagine trying to load 2,000 participants into 40 buses for a winery tour?

insideHPC: What do you think will be some of the hottest topics being discussed at ISC’10?

Wolfgang Gentzsch: HPC in the context of Multicore, Green IT, Cloud Computing and, certainly, the organizer of this conference for the past 25 years, Hans Meuer himself. And, purely from the networking standpoint, what are all my HPC friends doing this year?

insideHPC: While the ISC conference is on a much different scale than the ACM/IEEE SC conference, it is seen by many as being just as important — and in many ways — even more important for HPC companies looking to do business in Europe. How would you describe ISC to a company that is new to HPC?

Wolfgang Gentzsch: In such a fast changing area such as HPC, it makes a lot of sense to me to have two supercomputing conferences per year, every 6 months, one in the US and one in Europe. Change happens fast in this community — and once a year is just not enough to keep up with all the advancements. While the annual SC conference attracts a growing number of international attendees, there will always be some number of HPC professionals in Europe who just can’t do the travel to the annual US event. ISC brings a good balance and affords the vendors an opportunity to get close to their European customers.

ISC attracts various levels of decision makers in research, government and industry. So, for any HPC vendor, new or established, ISC is a great venue for meeting existing and future customers.

insideHPC: What advice do you have for first time attendees coming to ISC’10?

Wolfgang Gentzsch: Before you come to ISC, look at the rich online program. There is something of interest for everyone. My advice would be to make a detailed plan for the talks you wish to attend, the booths you would like to visit, the people you want to meet, and of course the parties where you can relax and network with your colleagues. Without some forethought and planning, the show can be overwhelming.

insideHPC: Tell us a little more about your role at DEISA — and can you explain the role of DEISA to our readers?

Wolfgang Gentzsch: DEISA is the Distributed European Infrastructure for Supercomputing Applications, providing access to HPC resources for individuals and teams of researchers, supporting them with solutions for solving their grand challenge big science applications. Hundreds of researchers in Europe and around the world have so far benefited from this Grid of HPC resources residing at 12 of the largest HPC centers in Europe. DEISA has now been in production for 5 years. My main role in DEISA is to work with the dissemination team, to spread the message widely, and to invite scientists and researchers to make use of this precious e-infrastructure.

insideHPC: Stepping back to take a much larger view of the global HPC community, what is your perspective on the HPC business climate for 2010 — and how does it compare to 2009?

Wolfgang Gentzsch: I would sum up the climate for 2010 as “evolution everywhere.” At one end (the smallest scale) we have increasing cores on the chip for lower power consumption, and at the other end (the large scale) we see the trend to consider Cloud services even for HPC applications. And in general, continuous commoditization with more user-friendly access, and more sectors using simulations on HPC systems.

So, how does this compare to the climate for 2009? I’d say it’s like a battery that’s been recharged and running at full speed. The activity level is exciting and HPC is experiencing renewed vigor and growth.

insideHPC: This brings up an interesting question. Some critics say we are putting way too much hype on Cloud services, and that it is really just a new label for Grid computing. What is your perspective on this?

Wolfgang Gentzsch: Cloud Computing is the result of a natural evolution of our community’s work on distributed computing. 10 years ago, when we talked about Grid Computing, our goal indeed was what Cloud Computing promises today: remote, secure, dependable, consistent, pervasive, and inexpensive access to computing, as just another (new) utility. That’s Cloud today. Grid remained more in the realm of the scientists, with their complex workflows which a Grid can accommodate, and with their needs to adjust (and match) applications and resources. And still, a Cloud can be a service node in a Grid, whenever a scientific workflow component is suitable for the simpler architecture of a Cloud.

Finally, in my opinion, Clouds will serve a useful purpose in a number of areas, but I don’t see Clouds, at least not in the midterm, for those environments where you need the highest performance computing while requiring low latency and powerful interconnect architectures for tightly coupled algorithms.

insideHPC: What do you see as the key enablers to us being able to really advance scientific discovery, and what are the key barriers?

Wolfgang Gentzsch: From my perspective, the most important key enablers will be easy access for every scientist to any HPC system, along with supporting interdisciplinary collaboration in virtual organizations.

A key barrier is that we still do not have enough scientists and engineers, compared with the challenges we face. Unfortunately, programs designed to attract many more students early on to science education is a topic still strongly neglected by many governments. I have some hope now with great technologies such as Web 2.0 and Cloud Computing, that we can start bringing science simulations and other content to our children and students, making it easy for them to access, and delivering programs in an “edutaining” way, such as what is being demonstrated already on the Web, for example, with the GridwiseTech e-School prototype and other projects.

insideHPC: Looking out to the next 3-5 years, how will HPC make a difference in our lives?

Wolfgang Gentzsch: HPC already makes a profound difference. Today, we are enjoying many amenities to make our lives easier, such as the advancements in travel, living, health, leisure, and knowledge.

Many advanced products and technologies have been designed on HPC systems, and HPC systems have allowed us to gain deeper insight into the secrets of nature. I have no doubt that this progress will continue and even accelerate in the coming years.

However, for HPC enthusiasts, this means: no hope for early retirement.

Wolfgang Gentzsch is currently Advisor to the EU funded project DEISA, a member of the Board of Directors of the OGF Open Grid Forum standards organization, and a senior consultant to HPC, Grid, and Cloud companies and governments. Before that he directed the German D-Grid Initiative and was an adjunct professor of computer science at Duke University in Durham and at NC State in Raleigh, and visiting scientist at the RENCI Renaissance Computing Institute at UNC Chapel Hill, North Carolina; Vice Chair of the EU e-Infrastructure Reflection Group e-IRG; and a member of the US President’s Council of Advisors for Science and Technology, PCAST.

Also posted in Events, Featured Stories, HPC People, ISC'10 Feature Stories, ISC10 | 1 Comment

Looking forward to ISC, an interview with conference founder Hans Meuer

An insideHPC Exclusive Interview with Hans Meuer, Co-founder and Organizer of the International Supercomputing Conference, ISC.

While the HPC community has seen its share of vendor companies that haven’t been able to demonstrate the staying power needed for survival, it’s quite the opposite story when it comes to conferences focused around this community.

The ACM/IEEE SC conference series will hold its 23rd annual meeting in New Orleans in November, the HPCC insider’s conference known as the Newport Conference has wraped up its 24th annual event in Newport, Rhode Island, and this year marks an impressive 25th anniversary for Europe’s most important HPC event, ISC’10, the International Supercomputing Conference.

insideHPC is pleased to share this exclusive interview with Prof. Dr. Hans Werner Meuer, the man behind the ISC conference.


insideHPC: First of all, congratulations on celebrating the 25th anniversary of the International Supercomputer Conference. It is truly a remarkable achievement. When you reflect on the history of this conference and how much you have achieved, what stands out as something you are particularly proud of?

Hans Meuer

Meuer: We started out in 1986 as a seminar at the University of Mannheim with 81 local attendees and 11 speakers. We are thus the HPC conference with the longest tradition worldwide. Over time we evolved to become the International Supercomputing Conference (ISC). For our 25th Anniversary in 2010, we are expecting a visitor crowd of 2,000 and as many as 140 exhibitors from all over the world. And, around 200 experts will be sharing their expertise at ISC’10.

We are proud to have realized early on that the IT world needs two major HPC events per year — the SC in the US alone is not sufficient to cater for such a fast-evolving sector like supercomputing. ISC is a good addition and also an alternative to the SC conference.

insideHPC: Will the difficult global economic situation have an impact on the conference this year, and what, if anything, are you doing differently this year because of the economy?

Meuer: Even during ISC’09, we were confronted with the global financial crisis and a poor economic situation. It was exactly one year ago that many of our sponsors predicted that our visitor numbers would plummet by 50 percent. On the contrary, our visitor numbers soared up by 21 percent. Even though there is of course no guarantee as to what things will be like this year, we will stick to our proven strategy. We have increased our efforts to extend and improve our overall concept and in particular our conference program and I believe we have succeeded — this year’s program is the best in our 25 years of history.

insideHPC: What do you believe will be some of the highlights of this year’s conference?

Meuer: Let me summarize the highlights of this year’s conference:

  • An absolute MUST will be High Performance Computational Life Sciences — The Challenge for HPC Systems
  • The panel about hitting the Exascale frontier in the future, including the keynote from Prof Dr. Horst, Zuse 80 Years of Computing: From Konrad Zuse to Exascale Computing.
  • The session Parallel Computing in the Years to Come focuses on the development of parallel computing tools, especially under the point of view of millions of cores.
  • The HPC-enabled simulation of global warming in the session Supercomputers for Modeling the Climate & the Roles of Energy Production in Climate Change
  • And the upcoming supercomputer countries in Evolving New HPC Markets: China, Middle East & Russia.

insideHPC: Do you have many first-time exhibitors coming to ISC10? Will the number of exhibitors be up from last year?

Meuer: This year we see a 20 percent surge in new exhibitors — about 30 businesses and scientific institutions will be exhibiting at ISC for the first time. We are hopeful to host around 140 exhibitors compared to 120 in 2009. Our exhibitors come from all continents, apart from Australia. For example, Brazil, South Africa, China, Japan, Russia, US and almost all European countries will be represented at our show.

Even though we increased our exhibition space by 15 percent, we are almost sold out at the beginning of February. There are only very few slots left in the industrial area.

Moreover, sponsors continue to value ISC as a core HPC event, which further contributes to the success of the exhibition.

insideHPC: Many people speak fondly about the networking events at ISC. How important are these networking events to the conference?

Meuer: The ISC networking events have become an integral part of our conference. They offer an effective instrument for establishing, maintaining and extending professional contacts. What participants appreciate most is the casual and relaxed atmosphere that makes it particularly easy to meet new and interesting people.

It is important to point out that these networking events are no longer limited to ISC itself but are now also taking place before and after ISC. When choosing the locations for our conferences, one crucial factor is to offer suitable networking locations also outside of the convention centers. Fortunately, with Hamburg we have found a location that offers ideal conditions.

As organizers, we would like to invite our participants and exhibitors to join our 25th Anniversary opening party on May 31.

insideHPC: What advice do you have for a first-time attendee to help them get the most out of ISC10?

Meuer: This year’s conference program offers a broad spectrum for all attendees. New attendees will benefit significantly from the seven tutorials on Sunday, May 30. Topics like GPUs, Manycore, Parallel Programming, Interconnects, LINPACK and HPC Software will be covered extensively in the tutorials.

For non-HPC experts whose day-to-day work demands HPC solutions, the Crash Course on High Performance Computing, on Wednesday, June 2 will prove to be beneficial.

And naturally, they shouldn’t miss the opening party, networking events and the Hot Seat session which is unique to ISC.

Finally, I would advise attendees to register before April 30 because the early bird fees will disappear forever after that.

insideHPC: I’ve heard some people refer to ISC as the “sister event” to the annual SC conference. But in fact, the two conferences are not linked in any way — is that correct?

Meuer: We are completely independent of the SC, even if we certainly acknowledge that the SC is the largest HPC event around. I think that we can both learn a lot from one another — and we surely make use of this opportunity. As for myself, I participated in all SC events so far — starting with Orlando in 1988 to Portland in 2009. Many HPC key players from the US have been ISC regulars from the very beginning of the conference, as speakers as well as attendees. As regards the involvement of the SC officials in ISC, I feel there is room for improvements. I believe, however, that this is due to a lack of continuity in the SC which is caused by the annual rotation of responsibilities.

Finally I take the opportunity to thank insideHPC for this wonderful opportunity.

See you soon in Hamburg!


For more background on Prof. Dr. Hans Werner Meuer, check out his CV [PDF].

insideHPC has already started running special feature coverage of ISC’10. To submit items for editorial coverage related to ISC’10, send us an email.

Also posted in Events, Featured Stories, HPC People, ISC'10 Feature Stories, ISC10 | 1 Comment

Advertisement


View All Videos

insideHPC.com is a production of insideHPC, LLC. © 2006-2011 Sitemap