“The next step is to look at using OpenMP directives to create multiple threads to distribute the work over many threads and cores. A key OpenMP directive, #pragma omp for collapse, will collapse the inner two loops into one. The developer can then set the number of threads and cores to use and return the application to determine the performance. In one test case, three threads per physical core shows the best performance, by quite a lot compared to just using one or two threads per core.”
The OpenPOWER Foundation is pleased to announce the first OpenPOWER Europe Summit taking place on June 16-17 in Frankfurt.
A new paper outlining NERSC’s Burst Buffer Early User Program and the center’s pioneering efforts in recent months to test drive the technology using real science applications on Cori Phase 1 has won the Best Paper award at this year’s Cray User Group (CUG) meeting.
Today D-Wave Systems announced the launch of Quantum for Quants, an online community designed specifically for quantitative analysts and other experts focused on complex problems in finance. Launched at the Global Derivatives Trading & Risk Management conference in Budapest, the online community will allow quantitative finance and quantum computing professionals to share ideas and insights regarding quantum technology and to explore its application to the finance industry. Through this community financial industry experts will also be granted access to quantum computing software tools, simulators, and other resources and expertise to explore the best ways to tackle the most difficult computational problems in finance using entirely new techniques.
In this video, Oklahoma State Director of HPC Dana Brunson describes how the Cowboy supercomputer powers research. “High performance computing is often used for simulations that may be too big, too small, too fast, too slow, too dangerous or too costly, another thing it’s used for involves data. So you may remember the human genome project it took nearly a decade and cost a billion dollars, these sorts of things can now be done over the weekend for under a thousand dollars. Our current super computer is named Cowboy and it was funded by a 2011 National Science Foundation Grant and it has been serving us very well.”
Today SGI announced the deployment of its largest SGI UV 300 supercomputer to date at The Genome Analysis Centre (TGAC) in the UK. As one of the largest Intel SSD for PCIe*deployments worldwide, TGAC’s new supercomputing platform gives the research Institute access to the next-generation of SGI UV technology for genomics. This will enable TGAC researchers to store, categorize and analyze more genomic data in less time for decoding living systems and answering crucial biological questions. “The combination of processor performance, memory capacity and one of the largest deployments of Intel SSD storage worldwide makes this a truly powerful computing platform for the life sciences.”
Today Fujitsu announced an order for a 25 Petaflop supercomputer system from the University of Tokyo and the University of Tsukuba. Powered by Intel Knights Landing processors, the “T2K Open Supercomputer” will be deployed at the Joint Center for Advanced High-Performance Computing (JCAHPC), which the two universities jointly operate. “The new supercomputer will be an x86 cluster system consisting of 8,208 of the latest FUJITSU Server PRIMERGY x86 servers running next-generation Intel Xeon Phi processors. Due to be completely operational in December 2016, the system is expected to be Japan’s highest-performance supercomputer.”
In this video from the 2016 OpenPOWER Summit, Stephen Bates of Microsemi presents: Enabling high-performance storage on OpenPOWER Systems. “Non-Volatile Memory (NVM), and the low latency access to storage it provides, is changing the compute stack. NVM Express is the de-facto protocol for communicating with local NVM attached over the PCIe interface. In this talk we will demonstrate performance data for extermely low-latency NVMe devices operating inside OpenPOWER systems. We will discuss the implications of this for applications like in-memory databases, analytics and cognitive computing. In addition we will present data on the emerging NVMe over Fabrics (NVMf) protocol running on OpenPOWER systems.”
The TERATEC Forum has posted their agenda for their 11th annual meeting. The event takes place June 28-29 in Palaiseau, France. “TERATEC brings together top international experts in high performance numerical design, simulation and Big Data, making it the major event in France and in Europe in this domain.”
Today the ISC Group, promoter of the TOP500 list, is pleased to announce the appointment of high performance computing industry journalist Michael Feldman to the position of managing editor of TOP500 News. In his new role, Feldman will expand the TOP500 project’s information portal into a comprehensive news site reporting on the HPC industry. “We are thrilled to bring Michael on board as he is a respected voice in the industry. With the addition of a comprehensive news site managed by him, the TOP500 project will offer a valuable new resource for HPC vendors and users,” said ISC Group managing director and TOP500 co-author, Martin Meuer.