Monica Ortiz writes "Panasas ActiveScale Storage Cluster Combined with Rocks Cluster Software Helps Maximize Linux Cluster Productivity and Eliminate I/O Bottlenecks.
FREMONT, Calif. (June 14, 2005) – Panasas, Inc., today announced that the Stanford University Institute for Computational and Mathematical Engineering (ICME) is utilizing the Panasas ActiveScale Storage Cluster in its high-performance computing (HPC) infrastructure. The Institute selected the Panasas Storage Cluster to work in conjunction with Rocks cluster software and eliminate the I/O bottleneck hindering overall cluster performance. Panasas will be hosting a free online seminar on Wednesday, June 29, 2005 at 11:00 AM Pacific Daylight Time to discuss how the Institute was able to maximize the utility of the cluster and accelerate results to its user community.
Researchers at the Institute use HPC as part of a comprehensive research program to better predict flutter and limit cycle oscillations for modern aircraft, better understand the impact of turbulent flow on jet engines, and to develop computational methodologies that can be expected to facilitate the design of naval systems with lower signature levels. They are using the 164-node Nivation Linux cluster to tackle large-scale simulations of flow/structure interaction, aeroacoustic, and elasto-acoustic scattering problems. Their research can lead to improvements in the performance and safety of aircraft, a reduction of engine noise that will improve the quality of life of people living near airports, and new underwater navigation systems.
“High-performance computing is critical to our success at the Institute. The productivity of our organization is dependent upon each cluster running at peak performance,” said Steve Jones, technology operations manager for the Institute. “By leveraging an object-based architecture, we are able to completely eliminate storage bottlenecks. Now we’re empowered to build larger, faster clusters, for we have a shared storage resource that can scale in both capacity and performance.”
The Institute relies on the object-based architecture delivered by Panasas to achieve high data bandwidth. The out-of-band Panasas DirectFLOW protocol enables direct node-to-disk access, resulting in as much as a 30X increase in data throughput for compute- and data-intensive applications. The Panasas Storage Cluster enables quick deployment by combining a next-generation, distributed file system with an integrated hardware/software solution that is fine-tuned to optimize performance. “We do science here, not IT,” said Jones. “We don’t have a huge team and weeks of time to integrate discrete systems. The solution was a breeze to deploy, taking only a few hours to set up.”
“The adoption of high-performance Linux clusters is creating a plethora of applications that have long-term implications on improving lives, ranging from gaining a better understanding of how jet engines perform to unlocking the genetic fabric of human cells in the race to prevent life-threatening diseases,” said Victor Perez, president and CEO at Panasas. “These applications are only possible by maximizing both the compute and storage power of Linux clusters.”
Steve Jones will be co-presenting with Larry Jones, vice president of product marketing at Panasas, during the “Rocks Clusters and Object Storage” webinar to be held on Wednesday, June 29, 2005 at 11:00 AM Pacific Daylight Time. To register, go to http://panasasevents.webex.com/ or send an email to Panasas_Events@panasas.com
The Rocks development community includes the Cluster Development Group at San Diego Supercomputer Center, Scalable Systems in Singapore, the HPC Group at University of Tromso in Norway, the SCE Group at Kasetsart University in Thailand, and the cluster development group at KISTI in Korea. Development of Rocks is funded from NSF, and aided by generous equipment donations from Sun Microsystems, Dell, AMD, Infinicon Systems, and Intel. The Rocks project was started at SDSC in early 2000, with the goal of "making clusters easy." Today, the Rocks user base includes five Top500 computers, and several hundred clusters around the globe. The Rocks Register, a web page where Rocks users voluntarily register their deployed cluster, shows that Rocks powers an aggregate of nearly 70 TFlops of peak computing. More information on Rocks, including documentation and complete access to the source code, can be found on the project's homepage (www.rocksclusters.org).
Panasas, Inc. helps companies accelerate the speed and accuracy of their business decisions, leading to real world breakthroughs that improve people’s lives. Panasas enables customers to maximize the benefits of Linux clusters by breaking down the storage bottleneck created by legacy network storage technologies. Through the delivery of the company’s Storage Cluster Platform, which combines industry-standard hardware with the company’s ActiveScale File System and professional services, the company has become the established leader in object-based, clustered storage. Panasas’ headquarters are in Fremont, CA with development facilities in Pittsburgh, PA and Houston, TX. For more information, please visit www.panasas.com.
# # #"