Stories about: Linux Networx

Silicon Graphics has rebounded enough from its financial woes to expand through acquisition, announcing last week that it has acquired the assets of high-performance cluster maker Linux Networx.

Once the domain of monolithic, multimillion-dollar supercomputers from Cray and IBM, HPC (high-performance computing) is now firmly within reach of today's enterprise, thanks to the affordable computing power of clustered standards-based Linux and Microsoft servers running commodity Intel Xeon and AMD Opteron processors. Many early movers are in fact already capitalizing on in-house HPC, assembling and managing small-scale clusters on their own.

Linux Networx, a high performance computing and clustering concern, recently delivered a pair of powerful Linux clustering products to two customers with eyes toward the skies - NASA's Center for Computational Sciences at the Goddard Space Flight Center, and ATK Launch Systems, a maker of weapons and space systems for the U.S. government.

ATK Launch Systems, a Brigham City, Utah-based provider of advanced weapons and space systems, has selected one of Linux Network's custom supercomputers, the companies said in a statement. The system is expected to increase ATK's compute capability to over 2.24 TFlops (trillion floating-pointoperations).

The NASA Center for Computational Sciences (NCCS) will get a new 3.3-TFLOP supercomputer from Linux Networx by mid-summer that will be used for continuing research on weather, climate variability and simulating astrophysical phenomena.

IBM on Tuesday announced that it was loosening its grip on its high performance General Parallel File System, opening the source code to third-party vendors so that the software could run on hardware other than its own.

Hardware and clustering company Linux Networx unveiled its LS Series of Linux machines Monday. The Series contains two families, the midrange LS-1 Supersystem and the high-end LS/X supercomputer.

Attendees at the recent BioIT World conference got a first look at a powerful new Linux clustering package for supercomputing environments.

To some extent, the whole point of using Linux clusters -- or any open-source platform -- for high-performance computing is to achieve consistently high compute speeds at modest price points. But when the speed falters, or the system's full performance potential isn't reached, the price point doesn't look like such a bargain.

The French Atomic Energy Commission (CEA) will be literally counting on Intel's forthcoming dual-core Itanium 2 processor when it takes delivery of a new supercomputer cluster next year. The Tera10 computing cluster will contain 4,352 of the chips, so far known only by their codename, Montecito, in 544 nodes.

As more companies explore the possibility of adding supercomputers to their IT stables, prices of the machines continue to fall even as their computing power and flexibility for business use is growing.