EARTH SCIENCES

SAN JOSE, Calif., ClusterWorld Conference & Expo -- DataDirect Networks' SwiftCluster was recognized by a distinguished panel of top industry professionals and the Linux Clusters Institute as the best storage clustering solution available today for GRID, HPC and HA computing. "The Excellence in Cluster Technology awards are the new industry standard for innovation in clustered systems," said Adam Goodman, CEO of QuarterPower Media, publisher of Linux Magazine and producer of ClusterWorld Conference & Expo. "These products and companies represent the very best solutions available to customers who require clustered systems. The developers should be applauded for their accomplishments." "This award validates our undisputed vision and leadership in storage clustering solutions for the HPC market," said Alex Bouzari, CEO of DataDirect Networks. "With dozens of the fastest HPC systems in the world getting significant performance, cost and scalability benefits from DataDirect Networks' products and services, we continue to provide the absolute best storage and clustering technology to our valued customers." "With SwiftCluster(TM) HPC customers are now able to build open and platform independent clusters scaling to tens of Teraflops with linear performance from 1 GB/Second to more than 10GB/Second in aggregate bandwidth," said Paul Bloch, President of DataDirect Networks. "Better yet, this best of breed technology is being deployed today at an unheard-of less than $1 Million per Teraflop, including client compute nodes, servers, interconnect and switching technology, storage and RAID controllers, disks and file systems." At the core of the SwiftCluster(TM) solution is DataDirect's S2A8000 Silicon Storage Appliance, the world's highest performance parallel storage controller. With its built-in distributed hardware RAID, a massive cache (up to 20GB), storage and LUN virtualization, parallel Fibre Channel port technology and support for hundreds of terabyte of FC or SATA storage, the S2A8000 is the enabling component behind some of the worlds leading HPC systems. A typical SwiftCluster(TM) solution, as the one demonstrated at ClusterWorld, is constructed from the following building blocks: * S2A8000 Appliances with virtualization capability * DMA engine that closely couples block level and file level elements of the cluster * High-performance switching infrastructure * Performance optimized TOE-enabled GigE host bus adapter * Standard computer node front end with off the shelf Intel Xeon or Itenium processors * Fibre Channel and SATA disk drives * Lustre global parallel file system