Storage comprises about 20% of the available budget in a typical HPC system, with another 20% being spent on software, 20% for services, and 40% on the compute cluster, according to marketing intelligence firm Intersect360 Research Firm (WW HPC Total Market Model). The majority of system design time and effort is spent validating the performance of the compute cluster, with less attention paid to storage. But not all storage is the same. All too frequently the investment and effort in building the best compute cluster will have yielded disappointing results when storage is not given...
By Barbara Murphy on August 31, 2011
Categories: Parallel Data Storage
By Barbara Murphy on August 23, 2011
Panasas captured headlines in 2008 when Los Alamos National Laboratory (LANL) deployed Panasas® ActiveStor® as the primary storage for Roadrunner, the world's first petascale supercomputer. Now, as institutions turn their sights to exascale computing, Panasas co-founder and CTO, Dr. Garth Gibson, explores the storage system capabilities that will be needed to deliver the required 70TB/s bandwidth while meeting the U.S. government’s 2018 target timeframe.
By Geoffrey Noer on August 13, 2011
It all started more than ten years ago when Panasas founder and CTO, Garth Gibson, realized that there was a strong emerging need for something faster than NFS to satisfy the growing I/O performance requirements of HPC Linux clusters. In 2004, Panasas started shipping its first ActiveStor® scale-out NAS solution running the PanFS® storage operating system. Key to the product’s success is the Panasas DirectFlow® protocol, the storage industry’s first implementation of...