arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1305.4650math.NAcs.NA
收藏

一种并行蝶形算法

A parallel butterfly algorithm

Jack Poulson, Laurent Demanet, Nicholas Maxwell, Lexing Ying

摘要:

蝶形算法是一种快速算法,当核K(x,y)在受限于满足某个简单几何条件的子域时近似为低秩时,该算法在大量目标点处近似计算积分变换\int K(x,y)g(y)dy的离散模拟。在d维空间中,当源点和目标点均为O(N^d)准均匀分布时,当K的每个适当子矩阵近似为rank-r时,算法的运行时间至多为O(r^2 N^d log N).本文介绍了一种蝶形算法的并行化方法,假定消息延迟为α,每个进程的带宽倒数为β,p个进程的执行时间最多为O(r^2 N^d/p log N + βr N^d/p + α)log p)。这种并行算法然后以开源DistButterfly库的形式实例化,用于K(x,y)=exp(i Φ(x,y))的特殊情况,其中Φ(x,y)是黑盒,足够平滑,实值相位函数。Blue Gene/Q上的实验证明了重要类相函数的令人印象深刻的强缩放结果。使用准均匀源,双曲Radon变换和一个类似的三维广义Radon变换分别观察到强规模从1节点/16核心到1024节点/16,384核心,效率分别大于90%和82%。

英文摘要:

The butterfly algorithm is a fast algorithm which approximately evaluates a discrete analogue of the integral transform \int K(x,y) g(y) dy at large numbers of target points when the kernel, K(x,y), is approximately low-rank when restricted to subdomains satisfying a certain simple geometric condition. In d dimensions with O(N^d) quasi-uniformly distributed source and target points, when each appropriate submatrix of K is approximately rank-r, the running time of the algorithm is at most O(r^2 N^d log N). A parallelization of the butterfly algorithm is introduced which, assuming a message latency of αand per-process inverse bandwidth of β, executes in at most O(r^2 N^d/p log N + βr N^d/p + α)log p) time using p processes. This parallel algorithm was then instantiated in the form of the open-source DistButterfly library for the special case where K(x,y)=exp(i Φ(x,y)), where Φ(x,y) is a black-box, sufficiently smooth, real-valued phase function. Experiments on Blue Gene/Q demonstrate impressive strong-scaling results for important classes of phase functions. Using quasi-uniform sources, hyperbolic Radon transforms and an analogue of a 3D generalized Radon transform were respectively observed to strong-scale from 1-node/16-cores up to 1024-nodes/16,384-cores with greater than 90% and 82% efficiency, respectively.

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。

↑