Hardware Technologies for High-Performance Data-Intensive Computing.

The article offers an investigation into hardware platforms suitable for data-intensive systems. It evaluates the benefits of two coprocessor architectures: graphics processors and reconfigurable hardware. It describes the Large Synoptic Survey Telescope (LSST) and the image-processing benchmark, th...

Full description

Bibliographic Details
Published in:	Computer (00189162) Vol. 41; no. 4; pp. 60 - 69
Main Authors:	Gokhale, Maya, Cohen, Jonathan, Yoo, Andy, Miller, W. Marcus, Jacob, Arpith, Ulmer, Craig, Pearce, Roger
Format:	Article
Published:	IEEE Apr2008
Subjects:	Computer input-output equipment Coprocessors Microprocessors Computer architecture Image processing Lanczos method Computer hardware description languages
Online Access:	View this record in EBSCOhost

Description
Summary:	The article offers an investigation into hardware platforms suitable for data-intensive systems. It evaluates the benefits of two coprocessor architectures: graphics processors and reconfigurable hardware. It describes the Large Synoptic Survey Telescope (LSST) and the image-processing benchmark, the Lanczos resampling filter. To assess the potential for accelerating LSST image-processing tasks, the authors' used the Lanczos filter. A key step in processing large document streams is language classification.