arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1008.1371math.NAcs.NA
收藏

基于GPU的双曲奇异值分解算法

A GPU-based hyperbolic SVD algorithm

Faculty of Mechanical Engineering and Naval Architecture, University of Zagreb(萨格勒布大学机械与船舶工程学院)

Vedran Novakovic, Sanja Singer

AI总结:

提出一种基于GPU的并行单边Jacobi双曲奇异值分解算法,用于对称不定特征值问题的最后阶段,相比串行和MPI并行变体在速度和精度上均有提升。

摘要:

开发了一种使用大规模并行图形处理单元(GPU)的单边Jacobi双曲奇异值分解(HSVD)算法。该算法也作为求解对称不定特征值问题的最后阶段。数值测试表明,与类似的Jacobi型HSVD算法的串行和MPI并行变体相比,该算法在速度和精度上均有提升。最后,讨论了混合CPU-GPU并行性的可能性。

英文摘要:

A one-sided Jacobi hyperbolic singular value decomposition (HSVD) algorithm, using a massively parallel graphics processing unit (GPU), is developed. The algorithm also serves as the final stage of solving a symmetric indefinite eigenvalue problem. Numerical testing demonstrates the gains in speed and accuracy over sequential and MPI-parallelized variants of similar Jacobi-type HSVD algorithms. Finally, possibilities of hybrid CPU--GPU parallelism are discussed.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑