Computerworld - Even in the sprawling data centers of Los Alamos National Laboratory, space is an issue. That makes blade servers an ideal approach to getting a big computing bang in a small space.
The New Mexico-based lab used System 324 blade servers from RLX Technologies Inc. in The Woodlands, Texas. Team leader Wu-chun Feng used the RLX blade server to upgrade the hardware for a 10 million particle simulation of the universe's formation. "Our 240-node server blade solution, code-named Green Destiny, replaced our 64-node, 128-processor VA Linux AltaCluster," says Feng. The same application runs in one-ninth the physical space inside the lab's data center.
Another key advantage of the blades for Los Alamos was the time it took to configure a full rack. Once a member of Feng's staff configured one blade in the 24-blade rack, that system image was simply added to the other 23 computers. The whole process took about two hours, he says.
Reliability has also been good. Feng says the Unix server cluster currently at the labs "is extraordinarily unreliable," failing about once a month. By contrast, the RLX server has been up for several months with no outages, he says.
Read more about Servers in Computerworld's Servers Topic Center.
Free Insider GuideCopyright © 1994 - 2012 Computerworld Inc. All rights reserved. Reproduction in whole or in part in any form or medium without express written permission of Computerworld Inc. is prohibited. Computerworld and Computerworld.com and the respective logos are trademarks of International Data Group Inc.
