June 21, 2011
Cray enters the International Supercomputing Conference (ISC) with a new generation of its flagship and midrange supercomputer lines, renewed momentum in Europe, and fresh perspectives on HPC market trends and technologies. HPCwire talked with Cray CEO Pete Ungaro and Ulla Thiel, vice president, EMEA, to get their perspective on the company's successes, challenges, and future plans.
HPCwire: What is Cray talking about this year at ISC?
Peter Ungaro: First off, we are very excited to be back at ISC again this year. For years, ISC has played a very important role in HPC, not only because it is a great HPC conference, but because it has also served to highlight Europe’s important role in the worldwide supercomputing community. It’s a great week, and we look forward to it every year.
The 2011 conference comes at an exciting time for us, as we just launched our new Cray XK6 and XK6m supercomputers at the Cray User Group meeting a few weeks back. The Cray XK6 is our first high-end GPU system, and it combines NVIDIA Tesla GPUs and the upcoming Interlagos processors from AMD with our own Gemini interconnect. We’re seeing a steady stream of visitors stopping by our booth and checking out the new the Cray XK6 blades, which we will have on display throughout the show. The system is scalable from a single cabinet to more than 50 petaflops, so it’s definitely worth a peek.
HPCwire: How important is Europe for Cray?
Ungaro: Well, we have a good run of contract wins in Europe over the last few years, so I would say Europe is really important for Cray! In all seriousness, Europe has been a big part of our recent growth and success as a company, with a number of large Cray systems going into customer sites throughout the continent. Just over the last few months, we’ve signed new contracts with leading European centers such as HLRS in Stuttgart, Germany, HECToR at the University of Edinburgh in Scotland, CSCS in Manno, Switzerland, Sweden’s Royal Institute of Technology in Stockholm, the University of Duisburg-Essen in Germany and a number of others.
It has been an incredible run of success for us, and we are working hard and doing our best to see that it continues. We’ve been hiring quite a bit in Europe across R&D, sales and service positions, and we still have a number of openings to fill. The HPC market in Europe is growing and it will continue to be a very important market for Cray well into the future.
HPCwire: What's the status of the plan to install a PRACE petaflop system at HLRS in Stuttgart? What will that system be used for?
Ulla Thiel: The HLRS petaflop system is scheduled to be installed in the fall. We were very excited to hear that this new Cray XE6 system at HLRS will be a PRACE Tier-0 system. It is one of three Tier-0 systems in Europe, of course bringing the latest and greatest technology to the table. The Cray system, which HLRS is already calling “HERMIT,” will also be the first Cray supercomputer outside the United States running at over a petaflops performance. Needless to say, we are very happy about our collaboration with HLRS and we are looking forward to seeing the great science that will be produced on this system within the HLRS and PRACE user communities.
HLRS has strong partnerships in the automotive and aerospace industries. As a member of the Gauss Center for Supercomputing, which is an alliance of the three major supercomputing centers in Germany, HERMIT will be a significant resource for users in the research and industrial HPC communities in Europe and beyond. Also, Cray and HLRS will undertake joint research activities in application areas within engineering sciences, which are also of interest to the automotive industry.
HPCwire: Is there anything different about the HPC market dynamics in Europe?
Thiel: The European HPC market is actually quite different from the HPC market in the U.S., most notably for its absence of a wide array of companies that develop supercomputers. However, Europe is very strong in the development of all kinds of software as well as in the applications space. Additionally, the European-wide HPC initiatives such as DEISA and PRACE have fostered a close collaboration between European countries. This collaboration has strengthened the industrial usage of supercomputing technologies and paved the way for future innovation, resulting in emerging technologies and economic exploitation in a variety of areas that have novel benefits for society.
HPCwire: You announced your Exascale Research Initiative at the end of 2009, with Edinburgh and Switzerland's CSCS as collaborators. How is that going?
Thiel: It’s going great, and we are very pleased with how the initiative continues to grow in terms of the scale and the number of projects it’s working on. Cray has made a big investment in R&D in Europe as part of this research initiative, which has allowed us to bring some excellent talent into the company. Exascale is big part of our R&D focus right now, and a lot of this work is being done in conjunction with our initiative in Europe, as well as through collaborations with other institutions around the world.
We like to believe that a Cray supercomputer will run an application with a sustained performance of over an exaflop by the end of the decade. It’s ambitious, we know, but with our significant R&D efforts in Europe and around the world, we’re doing everything we can to put a production exascale supercomputer into the hands of researchers, scientists and engineers.
HPCwire: What is Cray doing in Asia and how important is that market for you?
Ungaro: The Asia-Pacific market is another big growth area for Cray. We have had a number of successes and great customer partnerships in countries such as Japan, India, Australia and Korea, and we continue to expand into other countries. The earth sciences segment has been especially strong for Cray in Asia Pacific, and a great example of this is at the Korean Meteorological Administration, where their Cray XE6 is one of the largest operational weather forecasting systems in the world. We are pursuing new opportunities across this region, and we believe that it will continue to be an important part of our growth strategy in the future.
HPCwire: Can Cray break into the Chinese market?
Ungaro: One of the most exciting things I’ve seen over the last 20 years in HPC is the recent serious entrance of new countries into the supercomputing market, and that includes China. We have all heard about China’s recent supercomputing accomplishments and I must admit that they are quite impressive. China is demonstrating a real dedication to building large supercomputing systems and building computational sciences expertise. That being said, we believe there is a place for Cray supercomputers and technology in China. In fact, we are looking into establishing a subsidiary in Beijing and we hope to have something to announce on this soon. China has an interesting and expanding HPC marketplace and we believe we can add value there.
HPCwire: What role do GPUs play in Cray's strategy and for your customers?
Ungaro: One of the key elements of our new GPU systems, the Cray XK6 and XK6m, is the software environment we are developing -- and that’s where we truly believe the system will provide key benefits for customers looking to take advantage of GPUs. As we said at times during the launch, Cray isn’t the first to the GPU party, but we believe we have the best solution because the system has hardware that is fully integrated with software from Cray and from our partners. It’s a complete supercomputer that allows our customers to leverage both CPUs and GPUs by providing a single programming model across both environments. Customers can have a single version of their code running across both computing architectures.
GPUs will be an ongoing part of Cray supercomputers in the future, and we will continue to evolve our systems in line with our Adaptive Supercomputing vision, which means integrating diverse processing technologies into a unified architecture. We will look at other accelerator technologies in keeping with this vision, so long as they provide a benefit for our customers. We also plan on leveraging accelerators all the way to our exascale systems planned for the end of the decade. Even our Custom Engineering organization is offering some unique GPU configurations with different ratios of GPUs to CPUs than are currently found in our XK6 system.
Jun 18, 2013 |
The world's largest supercomputers, like Tianhe-2, are great at traditional, compute-intensive HPC workloads, such as simulating atomic decay or modeling tornados. But data-intensive applications--such as mining big data sets for connections--is a different sort of workload, and runs best on a different sort of computer.
Read more...
Jun 18, 2013 |
Researchers are finding innovative uses for Gordon, the 285 teraflop supercomputer housed at the San Diego Supercomputer Center (SDSC) that has a unique Flash-based storage system. Since going online, researchers have put the incredibly fast I/O to use on a wide variety of workloads, ranging from chemistry to political science.
Read more...
Jun 17, 2013 |
The advent of low-power mobile processors and cloud delivery models is changing the economics of computing. But just as an economy car is good at different things than a full size truck, an HPC workload still has certain computing demands that neither the fastest smartphone nor the most elastic cloud cluster can fulfill.
Read more...
Jun 14, 2013 |
For all the progress we've made in IT over the last 50 years, there's one area of life that has steadfastly eluded the grasp of computers: understanding human language. Now, researchers at the Texas Advanced Computing Center (TACC) are utilizing a Hadoop cluster on its Longhorn supercomputer to move the state of the art of language processing a little bit further.
Read more...
Jun 13, 2013 |
Titan, the Cray XK7 at the Oak Ridge National Lab that debuted last fall as the fastest supercomputer in the world with 17.59 petaflops of sustained computing power, will rely on its previous LINPACK test for the upcoming edition of the Top 500 list.
Read more...
05/10/2013 | Cleversafe, Cray, DDN, NetApp, & Panasas | From Wall Street to Hollywood, drug discovery to homeland security, companies and organizations of all sizes and stripes are coming face to face with the challenges – and opportunities – afforded by Big Data. Before anyone can utilize these extraordinary data repositories, however, they must first harness and manage their data stores, and do so utilizing technologies that underscore affordability, security, and scalability.
04/15/2013 | Bull | “50% of HPC users say their largest jobs scale to 120 cores or less.” How about yours? Are your codes ready to take advantage of today’s and tomorrow’s ultra-parallel HPC systems? Download this White Paper by Analysts Intersect360 Research to see what Bull and Intel’s Center for Excellence in Parallel Programming can do for your codes.
Join HPCwire Editor Nicole Hemsoth and Dr. David Bader from Georgia Tech as they take center stage on opening night at Atlanta's first Big Data Kick Off Week, filmed in front of a live audience. Nicole and David look at the evolution of HPC, today's big data challenges, discuss real world solutions, and reveal their predictions. Exactly what does the future holds for HPC?
Join our webinar to learn how IT managers can migrate to a more resilient, flexible and scalable solution that grows with the data center. Mellanox VMS is future-proof, efficient and brings significant CAPEX and OPEX savings. The VMS is available today.