PolyBench is a collection of benchmarks containing static control parts. The purpose is to uniformize the execution and monitoring of kernels, typically used in past and current publications.
We have modified the original PolyBench Suite to run on GPUs and accelerators by using different programming languages.
Any work that uses these codes should cite the following paper:
S. Grauer-Gray, L. Xu, R. Searles, S. Ayalasomayajula, J. Cavazos. "Auto-tuning a High-Level Language Targeted to GPU Codes." In Proceedings of Innovative Parallel Computing (InPar) 2012.
This work derives from the original PolyBench Suite created by Louis-Noël Pouchet.
This work was funded in part by:
U.S. National Science Foundation through the NSF Career award 0953667
DARPA Computer Science Study Group (CSSG)