Pavilion Data Systems will employ its HyperOS storage operating system with HyperParallel Flash Array to help Los Alamos National Laboratory offload analytics functions from storage servers and move them to the array in an effort to minimize data movement.
LANL looks to enable use of big data- or analytics community-made tools to perform data analytics by turning its file-based I/O into record or column based with help from Pavilion, the company said Wednesday.
DeltaFS technology is being used to allow LANL to use data reduction near the storage devices, demonstrating a thousand speedups on analytics functions.
“The file-based I/O had hidden the inherent structure in the data in those files. Switching to record/column-based I/O exposes the structure in the data, thereby enabling analytics,” said Gary Grider, High-Performance Computing Division leader at LANL.
He further noted, “Partnering with industry to explore this at extreme scale is the basis for this partnership with Pavilion.”