Data-driven analysis to understand GPU hardware resource usage of optimizations
- 5 March 2026
- journal article
- research article
- Published by SAGE Publications in The International Journal of High Performance Computing Applications
- Vol. 40 (3) , 404-420
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 in expensive accelerators. Although related work develops optimizations to improve application performance, none studies how these optimizations impact hardware resource usage or 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 multiobjective 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.Keywords
All Related Versions
Funding Information
- Office of Science (DE-SC0022843)
This publication has 20 references indexed in Scilit:
- Towards a Single-Host Many-GPU System2018
- PADDLE: Performance Analysis Using a Data-Driven Learning Environment2018
- Characterizing Power and Performance of GPU Memory Access2016
- GPU Computing Pipeline Inefficiencies and Optimization Opportunities in Heterogeneous CPU-GPU Processors2015
- Assessing a mini‐application as a performance proxy for a finite element method engineering applicationConcurrency and Computation: Practice and Experience, 2015
- PENNANT: an unstructured mesh mini‐app for advanced architecture researchConcurrency and Computation: Practice and Experience, 2014
- An investigation of Unified Memory Access performance in CUDA2014
- A quantitative performance analysis model for GPU architectures2011
- HPCTOOLKIT: tools for performance analysis of optimized parallel programsConcurrency and Computation: Practice and Experience, 2009
- Sparse non-negative solution of a linear system of equations is unique2008