arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1410.7249cs.DC
收藏

稀疏线性代数中可延展任务的调度树

Scheduling Trees of Malleable Tasks for Sparse Linear Algebra

Abdou Guermouche, Loris Marchal, Bertrand Simon, Frédéric Vivien

摘要:

科学工作负载通常被描述为有向非循环任务图。在本文中,我们专注于稀疏矩阵的多锋分解,其任务图被构造为一棵树的并行任务。在现有的并行任务模型中,可延展任务的概念特别强大,因为它允许每个任务在时变数量的处理器上处理。根据Prasanna和Musicus提出的矩阵计算模型,我们考虑了具有延展性的任务,其加速比为$p^α$,其中$p$是任务执行的处理器的分数份额,$α$($0 < α\leq 1$)是一个不依赖于任务的参数。我们首先激励我们的应用程序与实际实验在多核平台上的相关性,这个模型。然后,我们研究了由Prasanna和Musicus提出的最大完工时间最小化的最优控制理论的最优分配。我们在很大程度上简化了他们的证明,只诉诸纯粹的调度参数。基于这些新证据所获得的洞察力,我们将研究扩展到分布式多核平台。在那里,任务不能分布在多个分布式节点之间。在这样的分布设置(同构或异构),我们证明了相应的调度问题的NP-完全性,并提出了一些近似算法。最后,我们评估我们的方法的相关性模拟现实的树木。我们表明,我们的分配相对于现有的解决方案(因此不知道实际的加速功能)的平均性能增益是高达16%的$α=0.9$(在实际实验中观察到的值)。

英文摘要:

Scientific workloads are often described as directed acyclic task graphs. In this paper, we focus on the multifrontal factorization of sparse matrices, whose task graph is structured as a tree of parallel tasks. Among the existing models for parallel tasks, the concept of malleable tasks is especially powerful as it allows each task to be processed on a time-varying number of processors. Following the model advocated by Prasanna and Musicus for matrix computations, we consider malleable tasks whose speedup is $p^α$, where $p$ is the fractional share of processors on which a task executes, and $α$ ($0 < α\leq 1$) is a parameter which does not depend on the task. We first motivate the relevance of this model for our application with actual experiments on multicore platforms. Then, we study the optimal allocation proposed by Prasanna and Musicus for makespan minimization using optimal control theory. We largely simplify their proofs by resorting only to pure scheduling arguments. Building on the insight gained thanks to these new proofs, we extend the study to distributed multicore platforms. There, a task cannot be distributed among several distributed nodes. In such a distributed setting (homogeneous or heterogeneous), we prove the NP-completeness of the corresponding scheduling problem, and propose some approximation algorithms. We finally assess the relevance of our approach by simulations on realistic trees. We show that the average performance gain of our allocations with respect to existing solutions (that are thus unaware of the actual speedup functions) is up to 16% for $α=0.9$ (the value observed in the real experiments).

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。

↑