{"title":"Algorithm-based diskless checkpointing for fault tolerant matrix operations","authors":"J. Plank, Youngbae Kim, J. Dongarra","doi":"10.1109/FTCS.1995.466964","DOIUrl":null,"url":null,"abstract":"The paper is an exploration of diskless checkpointing for distributed scientific computations. With the widespread use of the \"network of workstations\" (NOW) platform for distributed computing, long-running scientific computations need to tolerate the changing and often faulty nature of NOW environments. We present high-performance implementations of several algorithms for distributed scientific computing, including Cholesky factorization, LU factorization, QR factorization, and preconditioned conjugate gradient. These implementations are able to run on PVM networks of at least N processors, and can complete with low overhead as long as any N processors remain functional. We discuss the details of how the algorithms are tuned for fault-tolerance, and present the performance results on a PVM network of SUN workstations, and on the IBM SP2.<<ETX>>","PeriodicalId":309075,"journal":{"name":"Twenty-Fifth International Symposium on Fault-Tolerant Computing. Digest of Papers","volume":"39 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1995-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"79","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Twenty-Fifth International Symposium on Fault-Tolerant Computing. Digest of Papers","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/FTCS.1995.466964","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 79
Abstract
The paper is an exploration of diskless checkpointing for distributed scientific computations. With the widespread use of the "network of workstations" (NOW) platform for distributed computing, long-running scientific computations need to tolerate the changing and often faulty nature of NOW environments. We present high-performance implementations of several algorithms for distributed scientific computing, including Cholesky factorization, LU factorization, QR factorization, and preconditioned conjugate gradient. These implementations are able to run on PVM networks of at least N processors, and can complete with low overhead as long as any N processors remain functional. We discuss the details of how the algorithms are tuned for fault-tolerance, and present the performance results on a PVM network of SUN workstations, and on the IBM SP2.<>
京公网安备 11010802042870号
京ICP备2023020795号-1
分享
求助内容：
应助结果提醒方式：
扫码关注我们
