Data-Driven Analysis to Understand GPU Hardware Resource Usage of Optimizations

TL;DR

This paper presents a data-driven analysis of GPU hardware resource usage, characterizing how optimizations impact utilization and performance, and identifying opportunities to enhance efficiency in scientific applications.

Contribution

It introduces a multi-objective metric for analyzing application-device interactions and demonstrates optimization strategies that improve GPU utilization and performance.

Findings

Optimizations can improve GPU utilization and execution time.

A multi-objective metric effectively identifies key application-device interactions.

Applying identified optimizations enhances performance and power efficiency.

Abstract

With heterogeneous systems, the number of GPUs per chip increases to provide computational capabilities for solving science at a nanoscopic scale. However, low utilization for single GPUs defies the need to invest more money for expensive ccelerators. While related work develops optimizations for improving application performance, none studies how these optimizations impact hardware resource usage or the average GPU utilization. This paper takes a data-driven analysis approach in addressing this gap by (1) characterizing how hardware resource usage affects device utilization, execution time, or both, (2) presenting a multi-objective metric to identify important application-device interactions that can be optimized to improve device utilization and application performance jointly, (3) studying hardware resource usage behaviors of several optimizations for a benchmark application, and finally (4) identifying optimization opportunities for several scientific proxy applications based on their hardware resource usage behaviors. Furthermore, we demonstrate the applicability of our methodology by applying the identified optimizations to a proxy application, which improves the execution time, device utilization and power consumption by up to 29.6%, 5.3% and 26.5% respectively.