Show simple item record

dc.contributor.authorPonce, Seanen_US
dc.contributor.authorJing, Huangen_US
dc.contributor.authorPark, Seung Inen_US
dc.contributor.authorKhoury, Chaseen_US
dc.contributor.authorQuek, Francisen_US
dc.contributor.authorCao, Yongen_US
dc.date.accessioned2013-06-19T14:35:53Z
dc.date.available2013-06-19T14:35:53Z
dc.date.issued2009-03-01
dc.identifierhttp://eprints.cs.vt.edu/archive/00001064/en_US
dc.identifier.urihttp://hdl.handle.net/10919/20162
dc.descriptionThis paper presents a novel parallelization and quantitative characterization of various optimization strategies for data-parallel computation on a graphics processing unit (GPU) using NVIDIA's new GPU programming framework, Compute Unified Device Architecture (CUDA). CUDA is an easy-to-use development framework that has drawn the attention of many different application areas looking for dramatic speed-ups in their code. However, the performance tradeoffs in CUDA are not yet fully understood, especially for data-parallel applications. Consequently, we study two fundamental mathematical operations that are common in many data-parallel applications: convolution and accumulation. Specifically, we profile and optimize the performance of these operations on a 128-core NVIDIA GPU. We then characterize the impact of these operations on a video-based motion-tracking algorithm called vector coherence mapping, which consists of a series of convolutions and dynamically weighted accumulations, and present a comparison of different implementations and their respective performance profiles.en_US
dc.format.mimetypeapplication/pdfen_US
dc.publisherDepartment of Computer Science, Virginia Polytechnic Institute & State Universityen_US
dc.subjectParallel computationen_US
dc.subjectAlgorithmsen_US
dc.subjectData structuresen_US
dc.titleAn Application-Oriented Approach for Accelerating Data-Parallel Computation with Graphics Processing Uniten_US
dc.typeTechnical reporten_US
dc.identifier.trnumberTR-09-05en_US
dc.type.dcmitypeTexten_US
dc.identifier.sourceurlhttp://eprints.cs.vt.edu/archive/00001064/01/paper.pdf


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record