September 18, 2008
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
Cray's new personal supercomputer;
New tool measures your HPC productivity against peers;
SGI delays reporting for fiscal year just ended;
Performing the Linpack on NCSA's #23 Win HPC Server machine;
Top500 launched in Asia;
Cluster Resources to Support Cray CX1;
Fujitsu Supercomputer Goes to History Museum;
HPC on Twitter;
Sun updates compute trailer;
TACC and the hurricane;
ClusterCorp Releases TotalView Rocks Roll;
>>McCain has answered Science Debate questions, too
McCain specifically calls out information technology research and computer science as important in a few of his answers. McCain says that he wants to invest in basic and applied research particularly in new and emerging areas and in information technology and will "support significant increases in basic research" at the various federal agencies -- but stopped short of saying he would fully fund the America COMPETES Act, in sharp contrast to Obama who has promised the doubling called for in that legislation.
You can (and voters should) read the whole thing at the Science Debate 2008 site, where they have placed McCain's responses alongside Obama's. As Melissa pointed out, and unlike Obama, McCain does refer to computers in his response: he uses the phrase "computer science" twice.
>>SiCortex Bumps Performance, Increases Energy Efficiency
SiCortex announced this week that it has doubled the price/performance ratio of its entire product line. The increase in efficiency comes by the way of a performance bump on the silicon, advancements in system software and leading-edge compilers. They released some initial TCO numbers compared to an Intel x86-based cluster.
"SiCortex has doubled its performance capacity without increasing the number of processors," said Christopher Stone, president and CEO of SiCortex. "Our computers now double the ROI at the time of purchase, and lower overall TCO by more than 60 percent over a three-year period."
For more info on the performance and efficiency increase, read the full release.
>>Appro partners with NEC
Appro has been in the news lately with wins in the DOE and Japan (University of Tsukuba). This week they announced a partnership with NEC. From the release:
Appro (http://www.appro.com), a leading provider of high-performance enterprise computing systems, and NEC Corporation, one of the world's leading providers of Internet, broadband network and enterprise business solutions, announce a strategic partnership today that will see Appro Xtreme-X Supercomputer and Appro Cluster Engine Software Management added and branded as part of NEC's HPC solution offering.
The partnership will enable Appro and NEC to work together towards a common goal; focus on reducing complexity of technology integration when deploying and managing integrated solutions, while lowering customer total cost of ownership (TCO). Commencing September 2008, Appro Supercomputer products will be added to NEC's HPC offering as a first step in this partnership.
Benefits to the participants? NEC gets a value-oriented solution and Appro's software (which is by all accounts good), and Appro gets access to the EMEA market where NEC is a much bigger presence than they are here in the U.S.
"This strategic partnership is a major breakthrough for Appro's supercomputers entry into the Europe, the Middle East and Africa (EMEA) HPC market. By combining NEC's strong technology base and market position in EMEA together with Appro's cluster deployment successes in the HPC market, the partnership will provide sustainable competitive advantages enabling both Appro and NEC to take greater advantage of this growing market segment," said Daniel Kim, CEO of Appro.
Although I wonder about the actual value of that access: as I read the announcement NEC will brand the Appro gear under NEC's own label. So while Appro will get revenue and deployment experience that will help them execute better on new business, they won't be directly growing their brand outside the US.
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?