Josef Weinbub

The performance portability of OpenCL kernel implementa-tions for common memory bandwidth limited linear algebra operations across different hardware generations of the same vendor as well as across vendors is studied. Certain combi-nations of kernel implementations and work sizes are found to exhibit good performance across compute kernels, hard-ware generations, and, to a lesser degree, vendors. As a consequence, it is demonstrated that the optimization of a single kernel is often sufficient to obtain good performance for a large class of more complicated operations