In this video, insideHPC interviews a special guest fake celebrity in the Panasas booth at ISC’12.
In this video, insideHPC interviews a special guest fake celebrity in the Panasas booth at ISC’12.
In this podcast, Jack Dongarra from the University of Tennessee discusses the June 2012 TOP500 list of the world’s most powerful supercomputers. The TOP500 is celebrating its 20th anniversary this year and Jack tells us more about how it got started and how the evolution of supercomputer performance measurement continues to evolve.
In this video, James Reinders discusses Allinea Software’s DDT debugger support for the newly announced Intel Xeon Phi (formerly known as Intel MIC) accelerator products. Recorded at ISC’12 in Hamburg.
We are pleased that Allinea Software has developed support for the Intel MIC Architecture and will be a valued resource for users of supercomputer systems using this exciting architecture” said James Reinders, Director of Parallel Programming Evangelism at Intel. “The first products from the new Intel® Xeon® Phi™ product family, based on Intel MIC Architecture, bring a new capability to supercomputer performance and power efficiency without sacrificing programmability by staying true to the flexibility of today’s Intel Xeon processor based systems. Allinea Software highlights this by extending their popular solutions in a manner that is completely familiar to users on today’s Intel Xeon processor based systems.”
Read the Full Story.
In this video, Josh Simons from VMware presents: Achieving Ultra-low Latency in the Cloud: How Low Can We Go? Recorded at the HPC Advisory Council Workshop at ISC’12 in Hamburg. Download the slides (PDF).
In this video, Doug Smith from the University of Colorado presents: HPC at the University of Colorado and the Student Cluster Competition. Recorded at the HPC Advisory Council Workshop at ISC’12 in Hamburg.
The United States has returned to a leading position on the TOP500 list with a new 16 Petaflop IBM BlueGene/Q supercomputer named Sequoia at LLNL. With 1,572,864 processing cores, Sequoia achieved an impressive 16.32 Petaflops on the Linpack benchmark to become the fastest machine on the planet.
In the number 2 position, Fujitsu’s K Computer weighs in at 10.51 Petaflops using 705,024 SPARC64 processing cores. The K Computer held the No. 1 spot on the previous two lists.
The new Mira supercomputer, an IBM BlueGene/Q system at Argonne National Laboratory in Illinois, debuted at No. 3, with 8.15 petaflop/s using 786,432 cores. The other U.S. system in the Top 10 is the upgraded Jaguar at Oak Ridge National Laboratory in Tennessee, which was the top U.S. system on the previous list and now clocks in at No. 6.
Coming in at number 4 on the list is SuperMUC at LRZ in Germany, an IBM IDataPlex cluster with innovative hot-water cooling. SuperMUC is now the fastest machine in Europe.
SuperMUC combines its hot-water cooling capability, which removes heat 4,000 times more efficiently than air, with 18,000 energy-efficient Intel Xeon processors. In addition to helping with scientific discovery, the integration of hot-water cooling and IBM application-oriented, dynamic systems management software, allows energy to be captured and reused to heat the buildings during the Winter on the sprawling Leibniz Supercomputing Centre campus, for savings of one million Euros ($1.25 million USD) per year.
Details on the rest of the TOP500 are still coming out here at ISC’12 in Hamburg, but I think it is worth noting that the TOP 20 systems on the list are all Petascale, a number that has doubled since November, 2011. Read the Full Story.
One of the most-anticipated demos coming up this week at ISC’12 will be from CAPS entreprise. During ISC 12, CAPS is going to demonstrate OpenACC portability by executing the same code on the leading massively parallel architectures: AMD Firestream, Intel MIC and Nvidia Tesla. With the first OpenACC deployments operated successively on these three platforms, directive-based programming reaches a key milestone.
We are very glad to demonstrate our latest achievements by making applications run on these various architectures with OpenACC. This unique code execution on different hardware is an illustration of our commitment to this standard.” says Eric Courtois, director of software development at CAPS.
You can check out the demos at the CAPS booth (#840) in Hamburg. Read the Full Story.
In this video, Torben Kling-Petersen presents an overview of Xyratex storage solutions for HPC.
Xyratex’s ClusterStor 3000 solution has generated considerable attention over the last year for its levels of throughput performance, density, and ease of management. This year, Xyratex announced the next generation in HPC storage with the new ClusterStor 6000, which significantly raises the bar on Lustre solutions with respect to performance, density, supportability, and manageability by enhancing the proven ClusterStor architecture, software stack, and increased data reliability.
Intersect360 Research recently launched the HPC500, an exclusive community of high performance computing user organizations that will share their real-world experiences to help shape the future of supercomputing. To find out more about the new organization and what’s coming up at ISC’12, I caught up with Intersect360 Research CEO Addison Snell.
insideHPC: How would you gage the overall health of the worldwide HPC market at this time?
Addison Snell: The good news is the worldwide HPC market is still growing, thanks to a strong commercial sector. Businesses worldwide are continuing to increase their investments in advanced technology for science, engineering, or business. Big Data is just one driver for this.
On the other hand, budget expectations for the public sector (academia and government) are the weakest we have seen them in five years, and this restricted growth—particularly in the supercomputer product segment—in 2011. The Eurozone crisis also caused some spending to slow down. Ultimately we still see long-term growth, but there will continue to be pockets of relative strength and weakness.
insideHPC: You recently announced the HPC500. What is that all about?
Addison Snell: The HPC500 is an exciting new bellwether group for the HPC industry. We’re of course looking for leading HPC organizations around the globe, but the HPC500 does more than that, by specifically gathering 500 organizations that form a representation of the industry across all vertical markets, geographies, and budget sizes.
User organizations will get a lot of benefits from participating, including no-cost access to our research and the ability to communicate with other HPC leaders worldwide. We’ve already admitted our first 50 members, and once we’re at full speed, we’ll have exclusive receptions and quarterly presentations on topics of interest. Our launch reception Monday night at ISC (beginning at 8:30 p.m. in the Trader Vic’s restaurant in the Radisson BLU) is our first event. And most of all, the HPC500 will be a guiding light for the industry, as their voices will inform our research and therefore the course of HPC development for our clients.
insideHPC: Do you have any predictions for the new TOP500 list?
Addison Snell: This is going to be a great year for TOP500 in its 20th anniversary. I think everyone wants to see whether K will repeat as number one. It will also be important to gauge how much momentum NVIDIA has with Tesla in advance of the introduction of Intel MIC systems.
But we also need to be careful to appreciate TOP500 for what it is: an example of what can be done by elite supercomputer users worldwide. These technologies and applications are awesome, and they inspire the rest of the industry to follow. But TOP500 is not about HPC market share. The broad industry doesn’t look like the TOP500 list. That’s where HPC500 can play a supportive and complementary role. We need to have both. I hope TOP500 sees another 20 years!
insideHPC: What activities does Intersect360 have planned for ISC’12?
Addison Snell: Our own reception for HPC500 on Monnday night will kick us off, and we look forward to doing similar events in the future. But beyond that, we are thrilled to be involved in so much of the program this year. Rupak Biswas of NASA and I will be the inquisitors for the first session of the newly revamped Hot Seat panel on Monday (1:00 to 3:00 p.m.), which now features shorter presentations and faster questions. And of course we wrap up the program on Thursday (1:00 to 2:00 p.m. in Hall 4) with the very popular ISC Analyst Crossfire panel. This year I’m thrilled to have as my panelists, Dona Crawford of LLNL, Happy Sithole of CHPC in South Africa, Elena Churakova of T-Platforms, and David Turek of IBM. It’s going to be fun!
insideHPC: What are your favorite aspects of ISC?
Addison Snell: The best thing about ISC is the international community building. Because of the leading role HPC has in society, I truly believe we are in a relationship business. It is such a benefit to stay in touch with all these great minds—and to meet new ones every year—from around the world, to see what’s new, what’s possible this year that you never saw before.
I love that the student cluster competition has such a big role at ISC this year. Attracting youth into HPC is going to be critical, not just for longevity of the industry, but for the fresh ideas they bring. I can’t wait to see what these students do!
insideHPC: How does ISC fit in the overall HPC ecosystem?
Addison Snell: There are other events and other forums, but ISC is the best one for making strong connections with colleagues old and new worldwide. It has a global reach to bring in the people I want to see from China, from Brazil, from Singapore, even from France!, but also plenty of opportunity to make those connections. The computers are always cool, but seeing what people are doing with them, that’s what keeps me addicted to the HPC industry and coming back to ISC year after year.
Recorded at ISC’12 in Hamburg.
Using Mellanox end-to-end InfiniBand solutions, SCINET has improved reliability and stability of their file systems, greatly improving the performance of parallel user jobs and user efficiency. SCINET is experiencing at least 15-20% increased performance out their upgraded cluster and expects to be high on the TOP500 list when the update is issued at ISC.
IDC is reporting that worldwide factory revenue for the HPC technical server market increased 3.1% in the first quarter of 2012 to reach $2.4 billion, up from $2.3 billion in the same period of 2011.
HPC technical servers, especially Supercomputers, have been closely linked not only to scientific advances but also to industrial innovation and economic competitiveness. For this reason, nations and regions across the world are increasing their investments in supercomputing even in today’s challenging economic conditions,” said Earl Joseph, program vice president for Technical Computing at IDC. “We expect the global race for HPC leadership in the petascale-exascale era to continue heating up during this decade.”
IDC expects the HPC technical server market to grow at a healthy 7.3% compound annual growth rate (CAGR) to reach revenues of $14 billion by 2016. Read the Full Story.
This week Eurotech announced that the newly formed European Technology Platform for HPC (ETP4HPC) will be meeting with customers and partners at ISC’12 in Hamburg. With a mission to improve Europe’s position in the domain of HPC technologies, vendors in the ETP4HPC include Allinea, ARM, Bull, Caps Entreprise, Eurotech, IBM, Intel, ParTec, STMicroelectronics, and Xyratex. Participating HPC research centers include BSC, CEA, CINECA, Fraunhofer, Forschungszentrum Juelich and LRZ.
The ETP is an industry led forum open to any organization performing research in the area of HPC technologies in Europe with the objective to define Europe’s research priorities in developing European technology in all segments of the HPC solution supply chain. The ETP strengthens European competitiveness in the area of HPC, which is a key requirement for future research and innovation as well as addressing a wide range of social and economic challenges.
insideHPC.com is a production of insideHPC, LLC. © 2006-2013 Sitemap