Platform Computing to Ease Grid DeploymentBy Jeffrey Burt | Posted 2006-09-26 Email Print
Re-Thinking HR: What Every CIO Needs to Know About Tomorrow's Workforce
The company is partnering with Dell and Intel to demo its Open Cluster Stack at IDF.
Platform Computing is looking to make it easier for businesses to create a high-performance computing cluster by offering an integrated reference software stack that can be quickly deployed and easily managed.
Platform is demonstrating its Open Cluster Stack, or OCS, at the Intel Developer Forum in San Francisco. The Platform OCS, which is running on a cluster of Dell computers, is being announced Sept. 26.
The offering is a modular software stack that integrates both open source and commercial software into a single cluster environment. Based on Platform's Rocks Cluster Toolkit, it gives customers an integrated and tested Linux cluster that is ready to deploy without much work on their part, said Gary Tyreman, vice president of the Open Cluster Group at Platform, of Markham, Ontario.
Tyreman said the company, which has been making a push into the enterprise with its grid products, has been watching over the past few years how businesses' buying habits have changed when creating clustered environments. For years, they would buy the components and then do the work of integrating them and building the cluster, he said.
"Now they're looking for the OEM to do it all," Tyreman said.
Platform has made a living selling cluster management components, and the time is right for the company to take the next step, he said. The enhanced software stack from the company helps customers avoid the higher development and sales costs that come with inconsistent Linux clusters.
Platform has worked closely with Dell for more than two years on a number of grid-related projects, and the work on OCS is a natural evolution of that relationship, Tyreman said.
At IDF, the demonstration will feature a number of off-the-shelf applications on a single software stack. Those applications include Fluent's computational fluid dynamics software and LS-DYNA from Livermore Software Technology.
"Our intention has been to build [the cluster stack] out and validate it," Tyreman said. "Now we're moving to the next level and communicating that to our customers."
The reference architecture is initially being offered on Dell infrastructures and using such technologies from Intel as its MPI runtime libraries.
Check out eWEEK.com's for the latest news, views and analysis on servers, switches and networking protocols for the enterprise and small businesses.