arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1708.08976cs.DC
收藏

稠密张量下MTTKRP的共享内存并行化

Shared Memory Parallelization of MTTKRP for Dense Tensors

Koby Hayashi, Grey Ballard, Jeffrey Jiang, Michael Tobia

摘要:

矩阵化张量乘Khatri-Rao积(MTTKRP)是计算张量CP分解算法的计算瓶颈。在本文中,我们开发了共享内存并行算法的MTTKRP涉及密集张量。该算法将几乎所有的计算都转换为矩阵运算,以便使用优化的BLAS子例程,并且它们避免了在内存中重新排序张量条目。我们的实现基准的顺序和并行性能,表现出高的顺序性能和高效的并行扩展。我们使用我们的并行实现来计算一个神经成像数据集的CP分解,并实现了高达$7.4\times$在现有的并行软件的加速比。

英文摘要:

The matricized-tensor times Khatri-Rao product (MTTKRP) is the computational bottleneck for algorithms computing CP decompositions of tensors. In this paper, we develop shared-memory parallel algorithms for MTTKRP involving dense tensors. The algorithms cast nearly all of the computation as matrix operations in order to use optimized BLAS subroutines, and they avoid reordering tensor entries in memory. We benchmark sequential and parallel performance of our implementations, demonstrating high sequential performance and efficient parallel scaling. We use our parallel implementation to compute a CP decomposition of a neuroimaging data set and achieve a speedup of up to $7.4\times$ over existing parallel software.

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。

↑