DRGPU: A Top-Down Profiler for GPU

Abstract—GPUs have become common in HPC systems to accelerate scientific computing and machine learning applications. Efficiently mapping these applications to rapidly evolutions of GPU architectures for high performance is a well-known challenge. Various performance inefficiencies exist in GPU kernels that impede applications from obtaining bare-metal performance. While existing tools are able to measure these inefficiencies, they mostly focus on data collection and presentation, requiring significant manual efforts to understand the root causes for actionable optimization. Thus, we develop DRGPU, a novel profiler that performs top-down analysis to guide GPU code optimization. As its salient feature, DRGPU leverages hardware performance counters available in commodity GPUs to quantify stall cycles, decompose them into various stall reasons, pinpoint root causes, and provide intuitive optimization guidance. With the help of DRGPU, we are able to analyze important GPU  benchmarks and applications and obtain nontrivial speedups — up to 1.77× on V100 and 2.03× on GTX 1650.