Linux NetworX will use HP's Itanium 2 servers as components in its systems, while HP will use the start-up's ClusterworX software to help manage supercomputers made of interconnected smaller systems. ClusterworX handles tasks such as installing software simultaneously on several computers, monitoring processor usage, and taking automatic actions such as shutting off a computer if a fan fails.
Supercomputing is a good market for Intel to try to secure customers for its still-new Itanium processor family, which works very differently from the established Celeron-Pentium-Xeon lineage and thus requires software to be rewritten. Supercomputer customers are willing to pay a premium for better mathematical abilities--which the Itanium possesses in spades--and often run their own programs so they aren't as dependent on outside software companies to support their needs.
HP has bet its entire server line on the Itanium family, which it helped to develop, while supercomputer niche companies such as Quadrics have begun supporting the new chips.
Supercomputers traditionally have been expensive, highly customized designs purchased by a select group of customers, but the industry is being overhauled by comparatively mainstream technologies such as Intel processors, InfiniBand high-speed connections, and Fibre Channel storage networks that have become fast enough to accomplish many tasks.
The new breed of supercomputers usually involve numerous two-processor servers bolted into racks and joined with special high-speed networks into a "cluster."
Linux NetworX customers include Los Alamos and Lawrence Livermore national laboratories for nuclear weapons research, Boeing for aeronautic engineering, and Sequenom for genetics research.
The new servers from Salt Lake City-based Linux NetworX will be available beginning in January, a representative said. The company will use both HP's two-processor rx2600 and its four-processor rx5670.
For its part, HP argues that using the Linux NetworX software will lower the overall cost to run supercomputing clusters.
