high performance computing on graphics processing units: hgpu.org

Hash aggregation is an important data processing primitive which can be significantly accelerated by modern graphics processors (GPUs). Previous work derived heuristics for GPU-accelerated hash aggregation from the study of a particular GPU. In this paper, we examine the influence of different execution parameters on GPUaccelerated hash aggregation on four NVIDIA and two AMD GPUs based on six different microarchitectures. While we are able to replicate some of the previous results, our main finding is that optimal execution parameters are highly GPU-dependent. Most importantly, execution parameters optimized for a specific GPU are up to 21× slower on other GPUs. Given this hardware dependency, we present an algorithm to optimize execution parameters at runtime. On GPUs with low runtime variation, our algorithm finds execution parameters that are less than 4% slower than the optimum on average and less than 18% slower in the worst case.