期刊文献+

一种基于日志合并优化的数据同步机制 被引量:6

A Data Synchronization Mechanism Based on Log-Merging Optimization
下载PDF
导出
摘要 基于复制技术的高可用存储系统,在故障恢复时需要进行数据同步操作以保证数据副本之间的一致,缩减数据同步操作的时间可减少系统修复时间,从而提高存储系统可用性.数据同步过程的时间由数据同步所需传输的数据量决定.本文提出的基于日志操作合并优化的数据同步机制充分考虑到文件访问中的时间和空间局部性,通过对故障期间的所记录的日志进行合并优化,合并和去除不必要的日志项,使得需要同步的数据量最少,从而减少数据同步过程的时间.试验的结果表明,相对于其它几种通用的方式,基于日志合并优化的数据同步机制所传输的数据量最少,可有效的减少数据同步操作的时间. For a high available storage system based on replication technology, the data of two mirrored sites must be synchronized while the system is being repaired after one of data replica crashed. Accelerating data synchronization process can cut down the whole time cost of system recovery, and therefore improves the availability of the storage system. The time cost of data synchronization much depends upon the quantity of the data which is moved in the process. In this paper,a new synchronization mechanism is present to reduce such quantity to a maximum extent. This mechanism based on log-merging optimization,which merges and deletes unnecessary file operations recorded in the log file before data being moved,consequently,only the modified data will be synchronized. And the result of experiment shows that ,this mechanism transfers the least quantity of data than other widely used mechanisms,As a result,it takes least time to complete the synchronization and then enhances the availability of whole system.
出处 《小型微型计算机系统》 CSCD 北大核心 2006年第12期2183-2188,共6页 Journal of Chinese Computer Systems
基金 国家"八六三"计划项目"新型网络服务器系统(一)"(2004AA111120)资助
关键词 复制 数据同步 日志合并优化 replication data synchronization log-merging optimization
  • 相关文献

参考文献9

  • 1Jim Senicka.Understanding veritas volume replicator[EB/OL].http://eval.veritas.com/mktginfoproducts/white-Papers/Storage-Server-Management/sf-vvr-wp.pdf,2005.
  • 2Ole Conradsen,Cyril Collin,Fabio Sabatti,etal.Disaster recovery using HAGEO and GeoRM[EB/OL].www.redbooks.ibm.com/redbooks/pdfs/sg242018.pdf,2000.
  • 3Mary G Baker,John H Hartmann,Michael D Kupfer,etal.Measurement of a distributed file system[C].Proceedings of the Thirteenth ACM Symposium on Operating Systems Principles,pages 198-212,Pacific Grove,CA,October 1991.
  • 4Werner Vogels.File system usage in windows NT 4.0[C].Proceedings of the Seventeenth ACM Symposium on Operating Systems Principles,Kiawah Island,SC,December 1999.
  • 5Nils Nieuwejaar,David Kotz,Apratim Purakayastha,etal.File-access characteristics of parallel scientific workloads[J].IEEE Transactions on Parallel and Distributed Systems,1996,7(10):.
  • 6Braam P,Callahan M,Schwan P.The intermezzo file system.O'Reilly PerlConference 3 Open Source Convention[C/OL].http://www.intermezzo.org/docs /perlinermezzo.pdf.August 1999.
  • 7James Jay Kistler.Disconnected operations in a distributed file system[D].Ph.D,Carnegie Mellon University,1993.
  • 8Carns P H,Ligon III W B,Ross R B,etal.PVFS:A parallel file system for linux clusters[EB/OL].http://www.parl.clemson.edu/pvfs/papers.html,2000.
  • 9Hui-I Hasio,David J Devitt.Chained declustering:a new availability strategy for multiprocessor database[C].Proceedings of Sixth International Conference on Data Engineering,Los Angel,CA,May,1990.

同被引文献43

  • 1侯孟书,卢显良,周旭,詹川.非结构化P2P系统复本研究[J].小型微型计算机系统,2005,26(11):1903-1906. 被引量:5
  • 2丁铖,张新华,梁正和.数字校园中数据同步方法的研究与实现[J].计算机工程,2006,32(20):270-272. 被引量:6
  • 3倪泳智,王珊.一种基于虚拟日志的数据复制解决方案[J].计算机科学,2007,34(3):78-82. 被引量:2
  • 4ABERER K, DATTA A, HAUSWIRTH M. Efficient, self- contained handling of identity in peer-to-peer systems[J]. IEEE Transaction on Knowledge and Data Engineering, 2004, 16(7): 858-869.
  • 5KALOGERAKI V, ZEINALIPOUR Y, Gunopulos D, et al. Distributed middleware architectures for scalable media services[J]. Journal of Network and Computer Applications, 2007, 30(1): 209-243.
  • 6CHOU J, HUANG Tai-yi. A scalable and load-balanced lookup protocol for high-performance peer-to-peer distributed systems[C]//Proceedings of International Conference on Parallel and Distributed Processing Techniques and Applications (PDPTA). Las Vegas, Nevada, USA: CSREA Press, 2003: 708-712.
  • 7KARAKAYA M, IBRAHIM K L, ULUSOY O Z. GnuSim: a general purpose simulator for GNUTELLA and unstructured P2P networks[R]. Bilkent University, 2005.
  • 8SAVITHA K, KARTHIKEYAN V, MARIO L. DiST: a scalable, efficient P2P lookup protocol[J]. Lecture Notes in Computer Science, 2005, 3601: 40-53.
  • 9LU Xi-cheng, ZHENG Qian-bing, ZHU Pei-dong, et al. An efficient long query path driven replication strategy in unstructured P2P network[C]//Proceedings of 4th International Conference on Networking (LNCS). [S. l. ]: Springer, 2005: 793-799.
  • 10WANG Yun, WANG Jun-ling. Load balancing framework for actively replicated servers[J]. Journal of Southeast University, 2005, 21(4): 419-426.

引证文献6

二级引证文献17

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部