期刊文献+

基于RS纠删码的HDFS云存储动态副本策略研究 被引量:2

Dynamic Replica Strategy in HDFS Cloud Storage based on RS Erasure Code
下载PDF
导出
摘要 目前多数云存储文件系统为保证其系统可靠性和性能所采用的静态完全副本冗余机制存在存储效率低、负载均衡能力差等缺点,为了提高云存储效率以及优化云存储的负载均衡能力,该文根据应用HDFS的经验,提出一种基于RS纠删码的云存储动态副本的冗余策略-SeaMaiden。SeaMaiden引入了副本调整模块和纠删码编解码模块,通过周期性的计算文件的热度,进而动态调整文件的副本数。当文件的副本数低于3时,则调用纠删码编解码模块对文件进行编码,提升文件的可靠性。当文件的热度级别上升时,则动态增加文件副本数,提升HDFS的负载均衡能力。通过实验证明SeaMaiden能有效提高HDFS的存储效率、可靠性及响应速度。 At present,in order to ensure reliability and performance of the storage file system,most cloud storage file systems used static redundancy mechanism.Its storage efficiency is low,and load balancing capability is poor.In order to improve the efficiency of cloud storage and to optimize the load balancing capability,based on the experience of HDFS,this paper presented SeaMaiden strategy which is a cloud storage dynamic replication strategy based on RS erasure code.SeaMaiden introduced the adjust copy module and erasure codes encoding and decoding module.By calculating the heat of the file periodically,HDFS can adjust the number of copies of documents.HDFS would call the encoding and decoding module to encode the file when the number of copies less than 3 to improve the reliability of the file.When the heat level of the file rose,HDFS would dynamically increase the number of the copies to improve the the load balancing capability.Finally,proved by experiments SeaMaiden can effectively improve the storage efficiency of HDFS,reliability and speed of response.
出处 《科技创新导报》 2013年第17期29-31,共3页 Science and Technology Innovation Herald
关键词 RS纠删码 动态副本策略 HDFS Hadoop Distributed File System(HDFS) Dynamic Replica Strategy RS
  • 相关文献

参考文献6

  • 1吴吉义,章剑林,傅建庆,平玲娣.基于Kademlia的云存储系统数据冗余方案研究[J].电信科学,2011,27(2):68-73. 被引量:17
  • 2程振东,栾钟治,孟由,李亮淑,和荣,杨婷婷,钱德沛,管刚,陈伟.云文件系统中纠删码技术的研究与实现[J].计算机科学与探索,2013,7(4):315-325. 被引量:9
  • 3http://hadoop.apache.org/docs/r2.0.4-alpha/hadoop-project-dist/hadoop-hd fs/HdfsDesign.html.
  • 4L.Rizzo,Effective erasure codes for reliable computer communication protocols,ACM Computer Communication,Review,1997,27(2).
  • 5WeatherspoonH,Kubiatowicz J.Erasure coding vs replication:quantitative comparison.In:Proc of the 1st Int'l Workshop Peer-to-Peer Systems,2002.
  • 6McKusick M K,Quinlan S.GFS:evolution onfast-forward[J].Queue,2009(7).

二级参考文献45

  • 1侯孟书,王晓斌,卢显良,任立勇.一种新的动态副本管理机制[J].计算机科学,2006,33(9):50-51. 被引量:13
  • 2田敬,代亚非.P2P持久存储研究[J].软件学报,2007,18(6):1379-1399. 被引量:52
  • 3Sanjay Ghemawat, Howard Gobioff, Shun-Tak Leung.The Google file system. In:Proc of the 19th ACM SOSP,New York,2003.
  • 4Kelly Sims. IBM introduces ready-to-use cloud computing collaboration services get clients started with cloud computing, http://www-03.ibm.com/press/us/en/pressrelease/22613.wss,2010.
  • 5Dhruba Borthaku.The hadoop distributed file system: architecture and design, http://hadoop.apache.org/common/docs/r0.16.0/hdfs_ design.pdf,2010.
  • 6Hbase Development Team. Hbase: bigtable-like structured storage for hadoop hdfs, http://wiki.apache.org/hadoop/Hbase,2010.
  • 7Amazon. Amazon simple storage service ($3), http://www.amazon. com/s3, 2010.
  • 8Kosmos file system (KFS), http://kosmosfs.sourceforge.net/,2010.
  • 9Gu Yunhong,Robert L Grossman. Sector and sphere: the design and implementation of a high-performance data cloud. Philosophical Transactions of the Royal Society,A(2009)367:2429-2445.
  • 10Robert L Grossman, Gu Yunhong. Data mining using high performance data clouds: experimental studies using sector and sphere. In: Proc of the 14th ACM SIGKDD,2008.

共引文献24

同被引文献8

引证文献2

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