GPA: A GPU Performance Advisor Based on Instruction Sampling

Developing efficient GPU kernels can be difficult because of the complexity of GPU architectures and programming models. Existing performance tools only provide coarse-grained tuning advice at the kernel level, if any. In this paper, we describe GPA, a performance advisor for NVIDIA GPUs that sugges...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:2021 IEEE/ACM International Symposium on Code Generation and Optimization (CGO) S. 115 - 125
Hauptverfasser: Zhou, Keren, Meng, Xiaozhu, Sai, Ryuichi, Mellor-Crummey, John
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 27.02.2021
Schlagworte:
Computer architecture
Graphics processing units
High performance computing
Kernel
Optimization
Parallel architectures
Parallel programming
Performance analysis
Programming
Tuning
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Developing efficient GPU kernels can be difficult because of the complexity of GPU architectures and programming models. Existing performance tools only provide coarse-grained tuning advice at the kernel level, if any. In this paper, we describe GPA, a performance advisor for NVIDIA GPUs that suggests potential code optimizations at a hierarchy of levels, including individual lines, loops, and functions. To relieve users of the burden of interpreting performance counters and analyzing bottlenecks, GPA uses data flow analysis to approximately attribute measured instruction stalls to their root causes and uses information about a program's structure and the GPU to match inefficiency patterns with optimization strategies. To quantify the potential benefits of each optimization strategy, we developed PC sampling-based performance models to estimate its speedup. Our experiments with benchmarks and applications show that GPA provides insightful reports to guide performance optimization. Using GPA, we obtained speedups on a Volta V100 GPU ranging from 1.01 x to 3.58 ×, with a geometric mean of 1.22 x.
DOI:10.1109/CGO51591.2021.9370339