并行水库模拟的预调节器

Proceedings of the International Conference on High Performance Compilation, Computing and Communications Pub Date : 2017-03-22 DOI:10.1145/3069593.3069618

Zhuo Tian, Bei Wang, Yifeng Chen

{"title":"并行水库模拟的预调节器","authors":"Zhuo Tian, Bei Wang, Yifeng Chen","doi":"10.1145/3069593.3069618","DOIUrl":null,"url":null,"abstract":"One of the standard preconditioners on reservoir simulation is incomplete LU (ILU, 'LU' stands for 'lower upper') decomposition. The matrix is factored by the product of a lower triangular matrix and an upper triangular matrix. Unfortunately, in parallel environments, ILU decomposition is serial for the solving process is Gaussian elimination which computes by rows. To enhance parallelism, we introduce an asynchronous parallel method. Asynchronous means that it is not need to wait other processes to finish, one process can begin to compute. In this way, ILU decomposition has a high degree of parallelism. Our test is based on reservoir simulation, in which the iterative algorithm is GMRES with ILU preconditioner. Experiment shows that the asynchronous parallel method can increase the parallelism of ILU decomposition and reduce the number of data exchanges.","PeriodicalId":383937,"journal":{"name":"Proceedings of the International Conference on High Performance Compilation, Computing and Communications","volume":"75 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2017-03-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":"{\"title\":\"Preconditioners for Parallel Reservoir Simulation\",\"authors\":\"Zhuo Tian, Bei Wang, Yifeng Chen\",\"doi\":\"10.1145/3069593.3069618\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"One of the standard preconditioners on reservoir simulation is incomplete LU (ILU, 'LU' stands for 'lower upper') decomposition. The matrix is factored by the product of a lower triangular matrix and an upper triangular matrix. Unfortunately, in parallel environments, ILU decomposition is serial for the solving process is Gaussian elimination which computes by rows. To enhance parallelism, we introduce an asynchronous parallel method. Asynchronous means that it is not need to wait other processes to finish, one process can begin to compute. In this way, ILU decomposition has a high degree of parallelism. Our test is based on reservoir simulation, in which the iterative algorithm is GMRES with ILU preconditioner. Experiment shows that the asynchronous parallel method can increase the parallelism of ILU decomposition and reduce the number of data exchanges.\",\"PeriodicalId\":383937,\"journal\":{\"name\":\"Proceedings of the International Conference on High Performance Compilation, Computing and Communications\",\"volume\":\"75 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2017-03-22\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"1\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Proceedings of the International Conference on High Performance Compilation, Computing and Communications\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1145/3069593.3069618\",\"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 International Conference on High Performance Compilation, Computing and Communications","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/3069593.3069618","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 1

摘要

油藏模拟的标准前提条件之一是不完全LU分解(ILU，“LU”代表“lower - upper”)。这个矩阵被分解成一个下三角矩阵和一个上三角矩阵的乘积。不幸的是，在并行环境中，ILU分解是串行的，因为求解过程是按行计算的高斯消去。为了提高并行性，我们引入了一种异步并行方法。异步意味着不需要等待其他进程完成，一个进程就可以开始计算。通过这种方式，ILU分解具有高度的并行性。我们的测试基于油藏模拟，其中迭代算法是带有ILU前置条件的GMRES。实验表明，异步并行方法可以提高ILU分解的并行性，减少数据交换次数。

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

Preconditioners for Parallel Reservoir Simulation

One of the standard preconditioners on reservoir simulation is incomplete LU (ILU, 'LU' stands for 'lower upper') decomposition. The matrix is factored by the product of a lower triangular matrix and an upper triangular matrix. Unfortunately, in parallel environments, ILU decomposition is serial for the solving process is Gaussian elimination which computes by rows. To enhance parallelism, we introduce an asynchronous parallel method. Asynchronous means that it is not need to wait other processes to finish, one process can begin to compute. In this way, ILU decomposition has a high degree of parallelism. Our test is based on reservoir simulation, in which the iterative algorithm is GMRES with ILU preconditioner. Experiment shows that the asynchronous parallel method can increase the parallelism of ILU decomposition and reduce the number of data exchanges.

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 International Conference on High Performance Compilation, Computing and Communications

自引率

0.00%

发文量