Linux Networx Announces Linux Supercomputing Storage Solutions

SALT LAKE CITY, Feb. 28 /PRNewswire/ -- Linux Networx, The Linux Supercomputing Company, today announced the availability of storage solutions optimized for a wide range of supercomputing needs. Integrating the appropriate storage configuration within a supercomputing system has been proven to power performance acceleration and utilization increases by a factor of 20 or more. Linux Networx storage solutions combine best-in-class hardware, management software and high performance file systems to best meet the unique requirements of each supercomputing environment. Linux Networx storage solutions are tightly integrated with the company's LS series of Linux Supersystems -- a family designed to deliver new levels of Total Application Throughput.

Drawing upon its extensive experience in Linux Supercomputing and applications optimization, Linux Networx has defined a range of storage solutions that deliver best-in-class storage to meet diverse supercomputing needs -- from sophisticated, extreme performance supercomputing applications to departmental workhorse production systems. Storage configurations are validated to deliver the optimum blend of price performance value, reliability and data availability demanded by each supercomputing class. The appropriate storage solution is then integrated and specifically tuned as part of a complete Linux Supersystem delivery. Leveraging Linux Networx GPFS, Linux Networx supercomputing solutions are easily integrated into existing performance computing environments.

"Linux Networx is committed to driving new levels of price/performance value through the delivery of application-optimized supercomputing systems. Integrated, optimized storage contributes significantly to overall supercomputing performance and utilization, the building blocks of total throughput and value," said Robert (Bo) H. Ewald, CEO of Linux Networx. "We're applying our real-world supercomputing experience to extend our Linux Supersystems with optimized storage solutions to drive enhanced performance and throughput, coupled with system-class reliability, ease of integration into existing environments and total cost of operations."

"A well-tuned I/O and storage subsystem is an essential element in delivering a high performance computing system," said Gene Poole, technical fellow, ANSYS. "Poor I/O performance is second only to inadequate main memory size in limiting many computing systems from delivery true high performance in our simulation software. A true high performance system must contain the appropriate, optimized disk capacity and throughput speed to feed today's Linux Supercomputing high speed processors. Failure to supply adequate memory or disk performance can increase solution time from 50-100 percent or more. By tightly integrating storage and compute capabilities, Linux Networx LS Series is moving to address this prevalent issue in the supercomputing arena."

As part of its commitment to delivering optimized supercomputing application performance, value and throughput, Linux Networx has tailored its storage offerings to meet the unique needs of three classes of supercomputing -- ultimate, premium and value. Ultimate performance environments include those conducting noise, vibration and harshness analysis in the automotive industry, managing national laboratory programs, or conducting seismic processing and structural analysis. For these users, compute intensive configurations must be complemented by an I/O and storage architecture that accesses and delivers data at extreme, sustained rates. These environments run very large, highly scalable single analysis jobs over extended periods of time. Reliable, timely execution of such critical supercomputing analysis is critical, demanding the ultimate levels of data availability and data integrity. For these users Linux Networx Ultimate Performance storage delivers the highest performance streaming bandwidth storage subsystems with industry-leading parallel file systems. The combination delivers best-of-breed performance and throughput coupled with superior on-line scalability and manageability.

Premium supercomputing environments require storage that can support multi-application environments with industry-standard software for applications such as overflow analysis in the aerospace market, computational fluid dynamics (CFD), process integration and design optimization, and reservoir simulation. For these users it is essential that the I/O and storage subsystem be easily configurable and reconfigurable to deliver the Total Application Throughput demanded by these blended application workloads. In addition, on-line capacity expansion is critical, as more users access the supercomputer to execute new applications. To meet these requirements Linux Networx provides enterprise-class storage management functionality, coupled with storage subsystems delivering flexible bandwidth and I/O performance capabilities, tightly integrated with industry leading parallel file systems. The result is a robust storage solution that delivers premium-class total throughput coupled with minimal total cost of operations.

Value supercomputing environments demand supercomputing power blended with simple management and deployment. Linux Networx delivers storage solutions featuring affordable I/O horsepower and best-in-class performance. NAS appliances simplify the storage/supersystem configuration and accelerate Time to Productivity. Typical Value Performance applications include small CFD and crash analyses.

Linux Networx storage solutions are available now as integrated and optimized subsystems within the company's LS series of Linux Supersystems as well as its Advanced Technology Clusters. LS series Supersystems blend application-optimized performance with a standardized systems experience to deliver ultimate supercomputing value, thereby delivering significantly reduced total cost of operations (TCO). About Linux Networx

For over a decade Linux Networx has delivered Linux Supercomputers specifically designed to drive new levels of Total Application Throughput for our customers. The company's Linux Supercomputing offerings power unrivaled value in application-optimized price/performance and utilization, overall system reliability, uptime and Total Cost of Operations. The company has delivered over 450 supercomputing clusters to more than 165 supercomputing customers around the world. In 2005, the company achieved record orders, finishing the year with three consecutive quarters of record growth. Our global customer base represents a "who's who" of supercomputing users ranging from Boeing in aerospace; BMW, Daimler Chrysler, Audi, Caterpillar and John Deere in heavy manufacturing; Total, Schlumberger and Shell Oil in the Oil and Gas industry; and Los Alamos National Laboratories, Sandia National Laboratories and the Department of Defense in the government arena. In 2005 the company was awarded three HPCWire Editor and Reader's Choice Awards -- for Most Innovative Technology in two categories and for Best Price/Performance. For more information about Linux Networx, visit http://www.linuxnetworx.com.