English

A Simple Model for Portable and Fast Prediction of Execution Time and Power Consumption of GPU Kernels

Distributed, Parallel, and Cluster Computing 2020-10-01 v3 Machine Learning Performance

Abstract

Characterizing compute kernel execution behavior on GPUs for efficient task scheduling is a non-trivial task. We address this with a simple model enabling portable and fast predictions among different GPUs using only hardware-independent features. This model is built based on random forests using 189 individual compute kernels from benchmarks such as Parboil, Rodinia, Polybench-GPU and SHOC. Evaluation of the model performance using cross-validation yields a median Mean Average Percentage Error (MAPE) of 8.86-52.00% and 1.84-2.94%, for time respectively power prediction across five different GPUs, while latency for a single prediction varies between 15 and 108 milliseconds.

Keywords

Cite

@article{arxiv.2001.07104,
  title  = {A Simple Model for Portable and Fast Prediction of Execution Time and Power Consumption of GPU Kernels},
  author = {Lorenz Braun and Sotirios Nikas and Chen Song and Vincent Heuveline and Holger Fröning},
  journal= {arXiv preprint arXiv:2001.07104},
  year   = {2020}
}