September 21, 2010
At the High Performance Computing Financial Markets conference in New York this week, Microsoft announced the third release of its Windows server operating system for technical computing, Windows HPC Server 2008 R2. And hot on the heels of that announcement were the usual slew of partner announcements from the likes of Cray, SGI, Adaptive Computing, Revolution Analytics, and Structured Data.
Writer Timothy Prickett Morgan covered the story for the Register, reporting on the strong positive response from the Wall Street crowd who were eager to try out the product. Many chose not to wait for the official release as evidenced by the several hundred users who participated in Microsoft's open beta program.
HPC Services for Excel 2010 is one of the highlights of the new release. It reduces calculation time of complex spreadsheets by dispatching the work from a Windows XP, Vista, or 7 workstation to a cluster of x64 servers running the lastest HPC Server. The feature will be especially attractive to the heavy users of complex spreadsheet models, the quantitative analysts (aka "quants") of Wall Street. Morgan noted that in a product demonstration given by Microsoft, a price sensitivity Excel workbook calculation that would have taken two hours to run on an eight-core Windows workstation was completed in under two minutes by offloading much of the calculation to a 500-node server cluster back at Microsoft headquarters.
Much coverage has been given to the cloud-bursting features of the lastest HPC Server variant and how this relates to Microsoft's Azure platform and overall cloud strategy. Writes Morgan:
At the event, Microsoft also showed a technology preview of a capability that is due in the next release of HPC Server 2008 that will allow clusters running the Microsoft supercomputing stack to burst applications out onto its Azure cloud. Adding Azure to the cluster will be no more difficult than plugging in your account information for Azure into the HPC Server console, feeding it a node template, and dispatching the software stack running on your internal cluster out to Azure. (This part takes time, obviously). But once the Azure nodes are configured with your software, they look and act the same as the nodes in your local cluster.
Ryan White, group program manager for high performance computing at Microsoft, says that the company will allow HPC Server 2008 to burst to other clouds, not just Azure. SGI's Cyclone cloud is on the list of future supported platforms, as is Amazon's EC2 cloud and Numbus Technology's Mezeo storage cloud.
Windows HPC Server 2008 R2 can also distribute computational workloads on Windows 7 PCs. This functionality is not supported on other Windows platforms (XP or Vista) or on Mac or Linux-based systems. Still, it's a neat program, comparable to a volunteer computing program, such as SETI@home. It can even be configured to stop processing when the user moves the mouse.
Full story at The Register
There are 0 discussion items posted.
|
Join the Discussion |
NVIDIA is telling everyone that the GK110, its new Kepler GPU aimed at supercomputing, is all about improving performance per watt. But the other driving theme behind the new architecture is reducing the GPU's reliance on its CPU host. How well it accomplishes both these goals areas could determine the success of the new chip in high performance computing.
Read more...
PGI, Cray, and CAPS enterprise are moving quickly to get their new OpenACC-supported compilers into the hands of GPGPU developers. At NVIDIA's GPU Technology Conference this week, there was plenty of discussion around the new HPC accelerator framework, and all three OpenACC compiler makers, as well as NVIDIA, were talking up the technology.
Read more...
NVIDIA has introduced its first Kepler-generation GPU product for high performance computing, and revealed some of the inner working of the new architecture. The announcement took place at the kickoff of the company's GPU Technology Conference taking place this week in San Jose, California.
Read more...