Work

DrGPUM: Guiding Memory Optimization for GPU-Accelerated Applications

Year: 2023

Type: conference-paper

Abstract: GPUs are widely used in today’s computing platforms to accelerate applications in various domains. However, scarce GPU memory resources are often the dominant limiting factor in strengthening the applicability of GPU computing. In this paper, we propose DrGPUM, the first profiler that systematically investigates patterns of memory inefficiencies in GPU-accelerated applications. The strength of DrGPUM, when compared to a large class of existing GPU profilers, is its ability to (1) correlate problematic memory usage with data objects and GPU APIs, (2) identify and categorize object-level and intra-object memory inefficiencies, and (3) provide rich insights to guide memory optimization.

Source: International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS)

DOI: 10.1145/3582016.3582044

Authors: Mao Lin, Keren Zhou, Pengfei Su

Institutions: University of California, Merced, OpenAI (United States)

Language: English

Topic: Parallel Computing and Optimization Techniques

Subfield: Hardware and Architecture

Field: Computer Science

Domain: Physical Sciences

Open Access status: hybrid

Funder: National Science Foundation

Award: Collaborative Research: CNS Core: SMALL: DrGPU: Optimizing GPU Programs via Novel Profiling Techniques