SC21 Proceedings

The International Conference for High Performance Computing, Networking, Storage, and Analysis

STM-Multifrontal QR: Streaming Task Mapping Multifrontal QR Factorization Empowered by GCN


Authors: Shengle Lin, Wangdong Yang, Haotian Wang, Qinyun Tsai, and Kenli Li (Hunan University)

Abstract: Multifrontal QR algorithm, which consists of symbolic analysis and numerical factorization, is a high-performance algorithm for orthogonal factorizing sparse matrix. In this work, a graph convolutional network (GCN) for adaptively selecting the optimal reordering algorithm is proposed in symbolic analysis. Using our GCN adaptive classifier, the average numerical factorization time is reduced by 20.78% compared with the default approach, and the additional memory overhead is approximately 4% higher than that of prior work. Moreover, for numerical factorization, an optimized tasks stream parallel processing strategy is proposed, and a more efficient computing task mapping framework for NUMA architecture is adopted in this paper, which is called STM-Multifrontal QR factorization. Numerical experiments on the TaiShan Server show average performance gains of 1.22x over the original SuiteSparseQR. Nearly 80% of datasets have achieved better performance compared with the MKL sparse QR on Intel Xeon 6248.


Presentation: file


Back to Technical Papers Archive Listing