A Performance Comparison of Sort and Scan Libraries for GPUs
Department of Computer Science, University of Cape Town, South Africa
University of Cape Town, 2014
Sorting and scanning are two fundamental primitives for constructing highly parallel algorithms. A number of libraries now provide implementations of these primitives for GPUs, but there is relatively little information about the performance of these implementations. We benchmark seven libraries for 32-bit integer scan and sort, and sorting 32-bit values by 32-bit integer keys.We show that there is a large variation in performance between the libraries, and that no one library has both optimal performance and portability.
October 20, 2014 by hgpu