Abstract
Graph neural networks (GNN) have shown great application potential in scientific research applications, biomedicine, and other fields, which exhibit superior feature representation capabilities for graph data with non-Euclidean structures. These capabilities are enabled efficiently by sparse matrix-matrix multiplication (SPMM) and sparse matrix-vector multiplication (SPMV) that operate on sparse matrix representations of graph structures. However, SpMM has the characteristics of high memory occupation and irregular memory access, which leads to low storage and computational efficiency. To address the above issues, this paper proposes a sparse matrix optimization method, including a sparse matrix format and a performance model. The format, namely BMCOO, divides the sparse matrix into multiple blocks and adopts the bitmap to compress the position information of non-zero elements in each block. This paper further designs an SpMV algorithm in BMCOO format on GPU. In addition, a multi-channel SpMV performance model is constructed to predict the execution time of SpMV by combining the sparse matrix scale and system architecture parameters. Then the performance model fine-tunes the graph partitioning of the GNN training process. Experiments on the SuiteSparse and the Open Graph Benchmark datasets verify the effectiveness and superiority of the proposed method.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Asanovic, K., et al.: The landscape of parallel computing research: a view from Berkeley. EECS Department, University of California, Berkeley, Tech. rep. (2006)
Davis, T.A., Hu, Y.: The university of Florida sparse matrix collection. ACM Trans. Math. Softw. (TOMS) 38(1), 1–25 (2011)
Hu, W., et al.: Open graph benchmark: datasets for machine learning on graphs. Adv. Neural. Inf. Process. Syst. 33, 22118–22133 (2020)
Huang, G., Dai, G., Wang, Y., Yang, H.: GE-SPMM: general-purpose sparse matrix-matrix multiplication on GPUs for graph neural networks. In: SC20: International Conference for High Performance Computing, Networking, Storage and Analysis, pp. 1–12. IEEE (2020)
Im, E.J., Yelick, K., Vuduc, R.: Sparsity: optimization framework for sparse matrix kernels. Int. J. High Perform. Comput. Appl. 18(1), 135–158 (2004)
Karakasis, V., Gkountouvas, T., Kourtis, K., Goumas, G., Koziris, N.: An extended compression format for the optimization of sparse matrix-vector multiplication. IEEE Trans. Parall. Distrib. Syst. 24(10), 1930–1940 (2012)
Karypis, G., Kumar, V.: A fast and high quality multilevel scheme for partitioning irregular graphs. SIAM J. Sci. Comput. 20(1), 359–392 (1998)
Liu, W., Vinter, B.: CSR5: an efficient storage format for cross-platform sparse matrix-vector multiplication. In: Proceedings of the 29th ACM on International Conference on Supercomputing, pp. 339–350 (2015)
Merrill, D., Garland, M.: Merge-based parallel sparse matrix-vector multiplication. In: SC’16: Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis, pp. 678–689. IEEE (2016)
Mu, Z., Tang, S., Zong, C., Yu, D., Zhuang, Y.: Graph neural networks meet with distributed graph partitioners and reconciliations. Neurocomputing 518, 408–417 (2023)
Wu, Z., Pan, S., Chen, F., Long, G., Zhang, C., Philip, S.Y.: A comprehensive survey on graph neural networks. IEEE Trans. Neural Netw. Learn. Syst. 32(1), 4–24 (2020)
Yan, S., Li, C., Zhang, Y., Zhou, H.: YASPMV: yet another SPMV framework on GPUs. Acm Sigplan Notices 49(8), 107–118 (2014)
Yao, T., et al.: Very short-term forecasting of distributed PV power using GSTANN. CSEE J. Power Energy Syst. (2022)
Zhao, Q., Yang, J., Wang, Z., Chu, Y., Shan, W., Tuhin, I.A.K.: Clustering massive-categories and complex documents via graph convolutional network. In: Qiu, H., Zhang, C., Fei, Z., Qiu, M., Kung, S.-Y. (eds.) KSEM 2021. LNCS (LNAI), vol. 12815, pp. 27–39. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-82136-4_3
Zhao, Y., Li, J., Liao, C., Shen, X.: Bridging the gap between deep learning and sparse matrix format selection. In: Proceedings of the 23rd ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming, pp. 94–108 (2018)
Acknowledgement
This work was supported by National Key R &D Program of China (No. 2021ZD0110403). We would like to thank the MindSpore team for their support.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Yao, T. et al. (2023). A Sparse Matrix Optimization Method for Graph Neural Networks Training. In: Jin, Z., Jiang, Y., Buchmann, R.A., Bi, Y., Ghiran, AM., Ma, W. (eds) Knowledge Science, Engineering and Management. KSEM 2023. Lecture Notes in Computer Science(), vol 14117. Springer, Cham. https://doi.org/10.1007/978-3-031-40283-8_11
Download citation
DOI: https://doi.org/10.1007/978-3-031-40283-8_11
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-40282-1
Online ISBN: 978-3-031-40283-8
eBook Packages: Computer ScienceComputer Science (R0)