稀疏迭代线性求解的低熵数据映射

Proceedings of the Conference on Extreme Science and Engineering Discovery Environment: Gateway to Discovery Pub Date : 2013-07-22 DOI:10.1145/2484762.2484797

M. Esmaily-Moghadam, Y. Bazilevs, A. Marsden

{"title":"稀疏迭代线性求解的低熵数据映射","authors":"M. Esmaily-Moghadam, Y. Bazilevs, A. Marsden","doi":"10.1145/2484762.2484797","DOIUrl":null,"url":null,"abstract":"An efficient parallel data structure implementation is presented to modify the permutation on the residual vector to achieve optimized memory layout of partitioned meshes for solving sparse linear systems. This novel algorithm is proposed to sort the data on each processor with respect to a set of rules. This simplifies implementation of parallel iterative solver algorithms and allows an overlap between non-blocking MPI communication and computations in matrix-vector product operations.","PeriodicalId":426819,"journal":{"name":"Proceedings of the Conference on Extreme Science and Engineering Discovery Environment: Gateway to Discovery","volume":"30 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2013-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"4","resultStr":"{\"title\":\"Low entropy data mapping for sparse iterative linear solvers\",\"authors\":\"M. Esmaily-Moghadam, Y. Bazilevs, A. Marsden\",\"doi\":\"10.1145/2484762.2484797\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"An efficient parallel data structure implementation is presented to modify the permutation on the residual vector to achieve optimized memory layout of partitioned meshes for solving sparse linear systems. This novel algorithm is proposed to sort the data on each processor with respect to a set of rules. This simplifies implementation of parallel iterative solver algorithms and allows an overlap between non-blocking MPI communication and computations in matrix-vector product operations.\",\"PeriodicalId\":426819,\"journal\":{\"name\":\"Proceedings of the Conference on Extreme Science and Engineering Discovery Environment: Gateway to Discovery\",\"volume\":\"30 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2013-07-22\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"4\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Proceedings of the Conference on Extreme Science and Engineering Discovery Environment: Gateway to Discovery\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1145/2484762.2484797\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the Conference on Extreme Science and Engineering Discovery Environment: Gateway to Discovery","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/2484762.2484797","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 4

摘要

提出了一种有效的并行数据结构实现，通过修改残差向量上的排列，实现了求解稀疏线性系统分区网格的优化内存布局。该算法根据一组规则对每个处理器上的数据进行排序。这简化了并行迭代求解算法的实现，并允许非阻塞MPI通信和矩阵向量乘积运算中的计算之间的重叠。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

Low entropy data mapping for sparse iterative linear solvers

An efficient parallel data structure implementation is presented to modify the permutation on the residual vector to achieve optimized memory layout of partitioned meshes for solving sparse linear systems. This novel algorithm is proposed to sort the data on each processor with respect to a set of rules. This simplifies implementation of parallel iterative solver algorithms and allows an overlap between non-blocking MPI communication and computations in matrix-vector product operations.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

Proceedings of the Conference on Extreme Science and Engineering Discovery Environment: Gateway to Discovery

自引率

0.00%

发文量