Machine and Application Aware Partitioning for Adaptive Mesh Refinement Applications

Machine and Application Aware Partitioning for Adaptive Mesh Refinement Applications
复制标题

自适应网格细化应用程序的机器和应用程序感知分区

DOI:
10.1145/3078597.3078610
复制
发表时间:
2017
期刊:
Proceedings of the 26th International Symposium on High-Performance Parallel and Distributed Computing
影响因子:
--
通讯作者:
Sundar, Hari
Sundar, Hari
中科院分区:
--
文献类型:
--
作者:
Fernando, Milinda;Duplyakin, Dmitry;Sundar, Hari

文献摘要

参考文献

被引文献

相似文献

在并行计算中,负载平衡和分区是至关重要的。基于空间填充曲线的流行分区策略侧重于平均分配工作。产生的分区独立于体系结构或应用程序。考虑到数据移动的相对成本不断增加,架构的异构性不断增加,仅仅考虑工作的平等划分已经不够了。最大限度地降低通信成本同样重要,如果不是更重要的话。我们的假设是,一个不平等的分区,最大限度地减少通信成本显着可以扩展和执行比传统的等工作分区计划。这种折衷取决于体系结构和应用程序。我们验证我们的假设的背景下,利用自适应网格细化的有限元计算。我们的核心贡献是一个新的分区方案,最大限度地减少后续计算的整体运行时间,通过执行架构和应用程序感知的非均匀工作分配,以减少解决方案的时间,主要是通过最大限度地减少数据移动。我们评估我们的算法进行比较,它对标准的空间填充曲线为基础的分区算法和观察时间的解决方案,以及解决自适应细化网格有限元计算的能量。我们证明了我们的新分区算法的良好的可扩展性高达tocores在ORNL的泰坦,并表明,提出的分区方案减少了整体能源以及时间解决方案的应用程序代码高达22.0%
Load balancing and partitioning are critical when it comes to parallel computations. Popular partitioning strategies based on space filling curves focus on equally dividing work. The partitions produced are independent of the architecture or the application. Given the ever-increasing relative cost of data movement and increasing heterogeneity of our architectures, it is no longer sufficient to only consider an equal partitioning of work. Minimizing communication costs are equally if not more important. Our hypothesis is that an unequal partitioning that minimizes communication costs significantly can scale and perform better than conventional equal-work partitioning schemes. This tradeoff is dependent on the architecture as well as the application. We validate our hypothesis in the context of a finite-element computation utilizing adaptive mesh-refinement. Our central contribution is a new partitioning scheme that minimizes the overall runtime of subsequent computations by performing architecture and application-aware non-uniform work assignment in order to decrease time to solution, primarily by minimizing data-movement. We evaluate our algorithm by comparing it against standard space-filling curve based partitioning algorithms and observing time-to-solution as well as energy-to-solution for solving Finite Element computations on adaptively refined meshes. We demonstrate excellent scalability of our new partition algorithm up tocores on ORNL's Titan and demonstrate that the proposed partitioning scheme reduces overall energy as well as time-to-solution for application codes by up to 22.0%
基于空间填充曲线的分层数据结构偏微分方程缓存感知算法
DOI: 10.1137/040604078
发表时间: 2006
期刊: SIAM J. Sci. Comput.
影响因子: --
作者:
Frank Günther;M. Mehl;Markus Pögl;C. Zenger
通讯作者: C. Zenger
使用空间填充曲线进行计算重新排序
DOI: --
发表时间: 2005
期刊:
影响因子: --
作者:
G. Jin;J. Mellor
通讯作者: J. Mellor
HykSort:分布式内存架构上超立方体快速排序的新变体
DOI: --
发表时间: 2013
期刊: International Conference on Supercomputing
影响因子: --
作者:
H. Sundar;D. Malhotra;G. Biros
通讯作者: G. Biros
DOI: --
发表时间: 2006
期刊: --
影响因子: --
作者:
Laxmikant;Kale Sanjeev
通讯作者: Laxmikant;Kale Sanjeev
Samplesort:最小存储树排序的采样方法
DOI: --
发表时间: 1970
期刊: JACM
影响因子: --
作者:
W. D. Frazer;A. C. McKellar
通讯作者: A. C. McKellar