September 17, 2009
Here is a collection of highlights from this week's news stream as reported by HPCwire.
SGI Sets Records, Outperforms Competitors
Chi-Tech Selects IBM's High Speed Messaging Technology
ANSYS Releases HFSS 12.0
New Cray Super Cuts Job Times at UT Dallas
NextComputing Presents NextStream
ScaleMP Supports Xeon 5500-based Platforms
ORAU Seeking Proposals for HPC Grants
Wolfram Releases webMathematica 3
Dataram Launches Financial Services Group
Intel Announces Organizational Changes
LSU Launches High Speed Cluster
Solace's Messaging Platform Breaks the Microsecond Barrier
Pliant Technology Unveils Enterprise Flash Drive Family
Researchers Find Cheaper Parallel Processing Using Xbox
HPC-Flavored Power7 Chips
There's was a good wrap-up of the upcoming IBM Power7 processors over at IT Jungle, penned by the prolific Timothy Prickett Morgan. Some of the article is a rehash of the news that came out of the Hot Chips conference last month, but with additional details about chip speeds and configurations. In particular, it looks like IBM is planning to offer multiple flavors of Power7 depending on its target application, varying not only the core count and clock speed, but also the number of memory channels and the width of the SMP buses.
One version will be of special interest to HPC'ers. Says Morgan:
There is another variant of the Power7 chip that will not appear in standard machines and which seems to be destined for the "Blue Waters" massively parallel supercomputer being built by IBM for the National Center for Supercomputing Applications at the University of Illinois. This is a multichip module (MCM) that will be a standalone SMP node that has four entire Power7 chips, with all eight of their memory controllers activated, plunked into a single ceramic package. These chips will all sport three 16-byte local links that glue the chips together. And there is every reason to believe these chips, which were developed under the code-name "Q7," will end up running at higher clock speeds than the MCM packaging should allow.
The use of processor variants to match up silicon with applications appears to be proliferating. Both Intel and AMD are using this approach with their latest server chip architectures, and now IBM is following suit. That bodes well for high performance computing, since it allows chipmakers to better target the HPC niche without having to invest in a completely customized design.
And Speaking of Blue Waters...
For the past couple years, NCSA has been releasing details on the Blue Waters project, which, when it comes online in June of 2011, is expected to be the most powerful supercomputer in the world for open scientific research. It will be the first system of its kind to deliver sustained performance of more than one petaflop on a range of real-world science and engineering applications.
For those of us who are interested in following the project's progress, there was a video posted as part of NCSA's Video on Demand series, featuring Bob Fiedler, Technical Program Manager for Science and Engineering Applications with the Blue Waters Project. In this, the fifth episode of the Behind Blue Waters series, Bob discusses how developers can prepare their applications for sustained petascale performance on the Blue Waters system, which is considerably larger than the machines most users and developers are familiar with.
Specifically, Bob discusses two kinds of simulators for testing applications' ability to run at large scale. One from IBM simulates a single execution on a single chip and the other, called BigSim, simulates the parallel execution of an application. Combining these tools enables a full system prediction of how fast the code will run and allows developers to make modifications that are needed to enable the applications to use the full capability that Blue Waters will provide.
To view the video, go to http://www.ncsa.illinois.edu/News/Video/2009/bbw_fiedler.html and for a related article by NCSA Director Thom Dunning that explains more about getting applications ready for the large-scale system, visit http://www.ncsa.illinois.edu/News/Stories/DunningPRAC/.
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?