gSampler: General and Efficient GPU-based Graph Sampling for Graph Learning

Gong, Ping1,3,4; Liu, Renjie2,3; Mao, Zunyao2,3; Cai, Zhenkun3; Yan, Xiao2; Li, Cheng4; Wang, Minjie3; Li, Zhuozhao2

10.1145/3600006.3613168

2023

29th ACM Symposium on Operating Systems Principles (SOSP)

PROCEEDINGS OF THE TWENTY-NINTH ACM SYMPOSIUM ON OPERATING SYSTEMS PRINCIPLES, SOSP 2023

OCT 23-26, 2023

Meta,Koblenz,GERMANY

1601 Broadway, 10th Floor, NEW YORK, NY, UNITED STATES

ASSOC COMPUTING MACHINERY

Graph sampling prepares training samples for graph learning and can dominate the training time. Due to the increasing algorithm diversity and complexity, existing sampling frameworks are insufficient in the generality of expression and the efficiency of execution. To close this gap, we conduct a comprehensive study on 15 popular graph sampling algorithms to motivate the design of gSampler, a general and efficient GPU-based graph sampling framework. gSampler models graph sampling using a general 4-step Extract-Compute-Select-Finalize (ECSF) programming model, proposes a set of matrix-centric APIs that allow to easily express complex graph sampling algorithms, and incorporates a data-flow intermediate representation (IR) that translates high-level API codes for efficient GPU execution. We demonstrate that implementing graph sampling algorithms with gSampler is easy and intuitive. We also conduct extensive experiments with 7 algorithms, 4 graph datasets, and 2 hardware configurations. The results show that gSampler introduces sampling speedups of 1.14-32.7x and an average speedup of 6.54x, compared to state-of-the-art GPU-based graph sampling systems such as DGL, which translates into an overall time reduction of over 40% for graph learning. gSampler is open-source at https://tinyurl.com/29twthd4.

Graph Neural Network Graph Sampling Graph Learning Graphics Processing Unit

其他

英语

CPCI-S

National Natural Science Foundation of China["62141216","62172382","61832011"]

Computer Science

Computer Science, Hardware & Architecture ; Computer Science, Software Engineering ; Computer Science, Theory & Methods

WOS:001135072900036

Web of Science

1.Univ Sci & Technol China, Hefei, Peoples R China
2.Southern Univ Sci & Technol, Shenzhen, Peoples R China
3.AWS Shanghai Lab, Shanghai, Peoples R China
4.Hefei Comprehens Natl Sci Ctr, Inst Artificial Intelligence, Hefei, Peoples R China

Gong, Ping,Liu, Renjie,Mao, Zunyao,et al. gSampler: General and Efficient GPU-based Graph Sampling for Graph Learning[C]. 1601 Broadway, 10th Floor, NEW YORK, NY, UNITED STATES:ASSOC COMPUTING MACHINERY,2023.