News

Oil firm pioneers IBM supercomputing on-demand service

A London-based oil exploration organisation has become the first customer of IBM's new super computing on-demand service.

IBM formally launched the service today, but PGS Data Processing, a division of Petroleum Geo-Services ASA, has been using the service for the past month.

PGS Data Processing takes raw seismic data and develops sophisticated maps for energy companies seeking new sources of oil or gas, said Chris Usher, president of the company.

The on-demand service allows PGS to avoid making large purchases from hardware suppliers based on short-term needs and to accept data processing jobs from its customers without having to worry about purchasing additional processing capacity, he said.

"Right now, if you have an urgent requirement, or have to expand your business, you have to make a long-term requirement for capacity, which needs to be linked to an increase in your market to show a return. Downturns stick you with fixed capacity and fixed costs," he said.

IT departments are under pressure to show a return on their technology investments, which is hard to do when expensive supercomputers sit unused, he added.

PGS started processing a job over an IBM Linux cluster in December, running it remotely from its London headquarters, Usher said. While the company's proprietary software runs on both Unix and Linux, PGS chose the Linux-Intel cluster because it was more familiar with the reliability of that configuration, he said.

IBM is offering companies a choice of platforms, either a network of Unix servers running IBM's Power4 processors or a Linux cluster of rack servers based on Intel's Xeon processors.

The supercomputing on demand offering is targeted at organisations with cyclical processing or design cycles, such as energy and life science companies and movie studios, according to Dave Jursik, vice-president of Linux clusters at IBM.

Interested customers sign a contract with IBM that specifies the length of the project, and the amount of IBM's resources needed to process the job. IBM would not provide specific pricing information, which differs for each customer depending on the amount of time required to finish a project and the amount of service required, among other things, Jursik said.

The initial clusters are located at IBM's facility in New York, but IBM expects to roll this technology out to its data centres around the world as it gains experience delivering the service.

As the service evolves, IBM hopes to move it toward a true utility-based computing service, where companies access computing resources like flipping a light switch, he added.

Email Alerts

Register now to receive ComputerWeekly.com IT-related news, guides and more, delivered to your inbox.
By submitting you agree to receive email from TechTarget and its partners. If you reside outside of the United States, you consent to having your personal data transferred to and processed in the United States. Privacy
 

COMMENTS powered by Disqus  //  Commenting policy