October 05, 2007
Here's a collection of highlights, selected totally subjectively, from this week's HPC news stream as reported at insideHPC.com and HPCwire.
>>10 words and a link
SGI marks 25 years of service;
37 national grid projects team on European Grid design study;
Scalable Informatics offers 48 TB in 5U for under $1/GB;
Winners of NSF Science & Engineering Visualization Challenge announced;
Intel, Dr. Dobbs partner on concurrency webinars;
IBM releases GPFS version 3.2;
PV-WAVE releases version 9.0;
SDSC offers Star-P on their on demand HPC system;
>>Appro Wins Contract for 438Tflops of NNSA Compute
Appro just released details on a recent contract award to the National Nuclear Security Administration for 438Tflops worth of cluster compute. The TLCC07 program represents a multi-million dollar contract award for Appro. The award consists of 3,024 nodes/12,096 sockets/48,348 cores split between eight different clusters directed towards the three main NNSA research facilities, LLNL/LANL/Sandia. All the machines will rely on quad-core AMD processors and ConnectX Infiniband. From the release:
"The ability to provide a scalable unit in multiple scalable clusters to all three national Lab sites with a 30-50 percent lower TCO is of great value to NNSA's effort to apply high performance computing to time-urgent national security challenges," said Mark Seager, Advanced Computing Technology lead for LLNL.
Details at http://www.hpcwire.com/hpc/1812897.html.
>>Intel: single-core performance boost
c|net's News.com is reporting from the Gelato Itanium Conference and Expo in Singapore that Intel's principal engineer of Itanium Processor Architecture, Cameron McNairy, said on Tuesday:
"We have not abandoned single-core performance and we're looking to increase that with each generation (of the Itanium processor) as we go forward, some with greater transitions than others."
"There are some workloads that you just need to have single-core performance in order to carry and to get the job done, and those are the kinds of things we are targeting," he said. "Not everything can scale and, truth be told, if you don't address the single-core performance segment, you're going to miss out on some opportunities."
>>SiCortex on the road to Europe
HPC startup SiCortex announced earlier this week that it's signed an agreement to sell its gear in Europe. From the release:
SiCortex...announced that it has entered into an agreement with MEGWARE, a leading European high performance computing systems integrator, to sell SiCortex's family of ultra compact, high performance Linux systems to the European market. MEGWARE will distribute SiCortex's products in Germany, Austria, Switzerland, Poland and the Czech Republic.
>>NSF Cyber-Enabled Discovery and Innovation program
The NSF is looking for proposals in a new program called Cyber-Enabled Discovery and Innovation (CDI). From the release:
CDI research outcomes are expected to produce paradigm shifts in our understanding of a wide range of science and engineering phenomena and socio-technical innovations that create new wealth and enhance the national quality of life.
Funding for this first project will range from $26 to $52 million, NSF's commitment to the CDI initiative for fiscal year 2008, and is expected to grow $50 million in each of the next five years. With this investment, NSF wishes to attract researchers to create revolutionary science and engineering research outcomes made possible by innovations and advances in "computational thinking," defined comprehensively as computational concepts, methods, models, algorithms, and tools.
>>Fernbach and Cray awards announced
Recipients for the prestigious Sidney Fernbach and Semour Cray awards, both to be presented in ceremonies at SC, have been announced.
David Keyes of Columbia is first up with the Fernbach, given by the IEEE Computer Society for "innovative uses of high performance computing in problem solving" each year. From the release:
Dr. Keyes is world-renowned for contributions to "Newton-Krylov-Schwarz" methods for the efficient solution of nonlinear partial differential equations on high performance computers. ... They have been incorporated into open mathematical software libraries that have enabled hundreds of users to make efficient use of parallel computers, from small clusters to the world's largest computers.
Next up is Kenneth Batcher of Kent State University for the Seymour Cray Science and Engineering Award, given by the IEEE Computer Society for "innovative contributions to high performance computer systems that exemplify the creative spirit of Seymour Cray." A high honor indeed.
Batcher, a Professor of Computer Science at Kent State University, is being recognized for fundamental theoretical and practical contributions to massively parallel computation, which involve distributing jobs across thousands of processors. His work has involved parallel sorting algorithms, interconnection networks, and pioneering designs of the STARAN and MPP computers.
You can find more details, including pointers to the releases, by clicking over to http://insidehpc.com/2007/10/01/fernbach-and-cray-awards-announced/.
10/30/2013 | Cray, DDN, Mellanox, NetApp, ScaleMP, Supermicro, Xyratex | Creating data is easy… the challenge is getting it to the right place to make use of it. This paper discusses fresh solutions that can directly increase I/O efficiency, and the applications of these solutions to current, and new technology infrastructures.
10/01/2013 | IBM | A new trend is developing in the HPC space that is also affecting enterprise computing productivity with the arrival of “ultra-dense” hyper-scale servers.
Ken Claffey, SVP and General Manager at Xyratex, presents ClusterStor at the Vendor Showdown at ISC13 in Leipzig, Germany.
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?