面向OpenMP的混合检查点机制

计算机科学与探索 ›› 2007, Vol. 1 ›› Issue (2): 191-199.

面向OpenMP的混合检查点机制

黄春+,刘勇鹏,杨学军

国防科技大学计算机学院，长沙 410073

收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-08-20 发布日期:2007-08-20
通讯作者: 黄春

A new hybrid mechanism for Checkpoint/Restart in OpenMP programs

HUANG Chun+,LIU Yongpeng,YANG Xuejun

School of Computer， University of Defense Technology， Changsha 410073， China

Received:1900-01-01 Revised:1900-01-01 Online:2007-08-20 Published:2007-08-20
Contact: HUANG Chun

摘要/Abstract

摘要： 检查点/续算是软件容错的重要途径之一。论文描述了一个系统级和应用级混合的OpenMP检查点机制，系统级支持不仅使检查点系统具有了好的透明性，并且使共享数据的保存不再由主线程单独完成，具有良好的数据局部性。应用级OpenMP协议将与OpenMP相关的协议处理独立出来，提高了系统的可移植性。NPB3.2-OMP测试结果表明，检查点和续算所需要的时间开销小，能够满足大规模程序的实际需求。

关键词: OpenMP, 检查点/续算, 系统级和应用级协同

Abstract: Checkpoint/Restart is one of the important approaches for software fault-tolerance. In this paper， the system-level and application-level coordinated Checkpoint/Restart mechanisms for OpenMP programs are presented. The system-level support is introduced for transparency， and it makes shared data saved by all threads together. The semantics-related operations of OpenMP will be separated from and hence independent of low-level systems by the application-level OpenMP checkpoint protocol， which improves portability of the checkpoint system. Based on the presented mechanism， a CCRG OpenMP Checkpoint/Restart system has been implemented. The experiments， such as NPB3.2-OMP， show the overhead of checkpointing and restarting is so limited that the system can be used in large scale programs.

Key words: OpenMP, Checkpoint/Restart, system-level and application-level coordinated

黄春+,刘勇鹏,杨学军. 面向OpenMP的混合检查点机制[J]. 计算机科学与探索, 2007, 1(2): 191-199.

HUANG Chun+,LIU Yongpeng,YANG Xuejun. A new hybrid mechanism for Checkpoint/Restart in OpenMP programs[J]. Journal of Frontiers of Computer Science and Technology, 2007, 1(2): 191-199.

[1]	车永刚，张理论，王勇献，徐传福，程兴华. Intel 多核与集成众核上CFD程序的OpenMP 性能分析[J]. 计算机科学与探索, 2015, 9(10): 1153-1162.
[2]	王庆, 季振洲, 刘涛. 面向嵌入式多核的OpenMP 扩展方法[J]. 计算机科学与探索, 2011, 5(1): 81-86.