HP Clusters with Linux NetworX on Supercomputers

By Michael Singer (Send Email)
Posted Dec 11, 2002


Clustering servers into one massive computing system has become all the rage in the Linux community, but two companies Tuesday say they are looking to take the technology to the next level. The partnership between the two companies centers on 64-bit Itanium 2-based systems running Linux for various clustering projects.

Hewlett-Packard said it is partnering with Salt Lake City-based Linux NetworX to expand each company's Linux cluster supercomputing offerings.

Linux NetworX designs clustering solutions for use in HPC and high availability (HA) applications. Palo Alto, Calif.-based HP and Linux NetworX have both designed and built multiple TeraFLOP Linux clusters, systems that are capable of performing trillions of calculations per second.

The computer and printer maker said it would offer Linux NetworX ClusterWorX software for preferred use with its HP ProLiant server line as a cluster supercomputing management tool. Linux NetworX, in turn, will offer HP Itanium 2-based systems running Linux. The idea is to give customers who want 64-bit capability on Intel chips without having to spend a fortune. The partnership does not include Linux NetworX's Evolocity Linux cluster platform.

Linux NetworX said it expects to roll out its first HP-enhanced Linux cluster products at the beginning of the year. The systems are expected to run HP's two-processor rx2600 server and four-processor rx5670 servers.

"By making comprehensive Linux cluster tools like ClusterWorX more widely available to the HPC community, we believe this will further accelerate the adoption of Linux clusters - you no longer have to be an expert in Linux and clustering to take advantage of this industry-changing technology," said Linux NetworX president and CEO Stephen Hill.

The two companies say they are a natural fit for cluster networking on the Linux operating system. HP is the hardware vendor with the most systems on the Top500 supercomputer list, while Linux NetworX and HP have both designed and built clusters that rank among the ten fastest supercomputers in the world.

Case in point, the U.S. Department of Energy tapped Linux NetworX this year to build the world's fastest Linux supercomputer at its Lawrence Livermore National Laboratory for to support the lab's national security mission.

Linux NetworX is also in co-development with LLNL for SLURM (Simple Linux Utility for Resource Management). SLURM is an open source resource management system developed for Linux clusters that focuses on portability, interconnect independence, fault-tolerance and security.

HP has a long-standing relationship with the DOE, building a 9.2 teraflop Linux systems supercomputer for the Northwest National Laboratory (PNNL) in Richland, Washington.

Page 1 of 1


Comment and Contribute

Your name/nickname

Your email

(Maximum characters: 1200). You have characters left.