Compute Unified Device Architecture Application Suitability
Wen‐mei Hwu, Christopher I. Rodrigues, Shane Ryoo, John A. Stratton · Computing in Science & Engineering · 2009
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.