A Simple Model for Portable and Fast Prediction of Execution Time and Power Consumption of GPU Kernels
This work addresses efficient task scheduling for GPU users, but it is incremental as it builds on existing methods with a focus on portability and speed.
The paper tackles the problem of predicting GPU kernel execution time and power consumption by developing a simple, portable model using hardware-independent features, achieving median MAPE of 8.86-52.00% for time and 1.84-2.94% for power across five GPUs.
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.