Compute Unified Device Architecture Application Suitability

Compute Unified Device Architecture Application Suitability

Graphics processing units (GPUs) can provide excellent speedups on some, but not all, general-purpose workloads. Using a set of computational GPU kernels as examples, the authors show how to adapt kernels to utilize the architectural features of a GeForce 8800 GPU and what finally limits the achieva...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Computing in science & engineering 2009-05, Vol.11 (3), p.16-26
Hauptverfasser: Hwu, Wen-Mei, Rodrigues, Christopher, Ryoo, Shane, Stratton, John
Format: Artikel
Sprache:eng
Schlagworte:
Architecture
benchmarks
Central Processing Unit
Computation
compute unified device architecture
Computer architecture
Costs
CUDA
Devices
general-purpose computing on GPU
GPGPU
Graphics
Hardware
Kernel
Kernels
Multicore processing
Parallel processing
Phased arrays
software optimization
Workload
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Graphics processing units (GPUs) can provide excellent speedups on some, but not all, general-purpose workloads. Using a set of computational GPU kernels as examples, the authors show how to adapt kernels to utilize the architectural features of a GeForce 8800 GPU and what finally limits the achievable performance.
ISSN:1521-9615
1558-366X
DOI:10.1109/MCSE.2009.48