arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1607.00346math.NAcs.MScs.NA
收藏

分布式存储层次插值分解

Distributed-memory Hierarchical Interpolative Factorization

Yingzhou Li, Lexing Ying

摘要:

分层插值因子分解(HIF)为椭圆型偏微分方程的求解或预处理提供了一种有效的方法。通过利用算子的局部性和低秩性质,HIF实现了分解离散正定椭圆算子的准线性复杂度和求解相关线性系的线性复杂度.在本文中,分布式存储器HIF(DHIF)介绍了作为一个并行和分布式存储器实现的HIF。DHIF以分层结构组织进程,并尽可能保持通信的本地性。构造和应用DHIF的计算复杂度分别为$O\left(\frac{N\log N}{P}\right)$和$O\left(\frac{N}{P}\right)$,其中$N$是问题的大小,$P$是进程的数量。通信复杂度为$O\left(\sqrt{P}\log^3 P\right)α+ O\left(\frac{N^{2/3}}{\sqrt{P}}\right)β$,其中$α$为延迟,$β$为反向带宽。大量的数值例子进行NERSC爱迪生系统多达8192个进程。数值结果与复杂度分析一致,证明了DHIF的有效性和可扩展性。

英文摘要:

The hierarchical interpolative factorization (HIF) offers an efficient way for solving or preconditioning elliptic partial differential equations. By exploiting locality and low-rank properties of the operators, the HIF achieves quasi-linear complexity for factorizing the discrete positive definite elliptic operator and linear complexity for solving the associated linear system. In this paper, the distributed-memory HIF (DHIF) is introduced as a parallel and distributed-memory implementation of the HIF. The DHIF organizes the processes in a hierarchical structure and keep the communication as local as possible. The computation complexity is $O\left(\frac{N\log N}{P}\right)$ and $O\left(\frac{N}{P}\right)$ for constructing and applying the DHIF, respectively, where $N$ is the size of the problem and $P$ is the number of processes. The communication complexity is $O\left(\sqrt{P}\log^3 P\right)α+ O\left(\frac{N^{2/3}}{\sqrt{P}}\right)β$ where $α$ is the latency and $β$ is the inverse bandwidth. Extensive numerical examples are performed on the NERSC Edison system with up to 8192 processes. The numerical results agree with the complexity analysis and demonstrate the efficiency and scalability of the DHIF.

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。

↑