基于压缩近邻的查重元数据去冗算法设计

doi:10.11959/j.issn.1000-436x.2015226

通信学报 ›› 2015, Vol. 36 ›› Issue (8): 1-7.doi: 10.11959/j.issn.1000-436x.2015226

• 学术论文 • 下一篇

基于压缩近邻的查重元数据去冗算法设计

姚文斌^1,²,叶鹏迪³,李小勇⁴,常静坤^1,²

¹ 北京邮电大学智能通信软件与多媒体北京市重点实验室，北京 100876
² 北京邮电大学计算机学院，北京 100876
³ 中国铁道科学研究院机车车辆研究所，北京 100081
⁴ 北京邮电大学可信分布式计算与服务教育部重点实验室，北京 100876

出版日期:2015-08-25 发布日期:2015-08-25
基金资助:
国家自然科学基金资助项目;国家高技术研究发展计划（“863计划）基金资助项目;中央高校基本科研业务费专项基金资助项目

Deduplication algorithm based on condensed nearest neighbor rule for deduplication metadata

Wen-bin YAO^1,²,Peng-di YE³,Xiao-yong LI⁴,Jing-kun CHANG^1,²

¹ Beijing Key Laboratory of Intelligent Telecommunications Software and Multimedia,Beijing University of Posts and Telecommunications,Beijing 100876,China
² School of Computer Science,Beijing University of Posts and Telecommunications,Beijing 100876,China
³ The Locomotive and Car Research Institute,China Academy of Railway Sciences,Beijing 100081,China
⁴ Key Laboratory of Trustworthy Distributed Computing and Service of Ministry of Education,Beijing University of Posts and Telecommunications,Beijing 100876,China

Online:2015-08-25 Published:2015-08-25
Supported by:
The National Natural Science Foundation of China;The National High Technology Research and Development Program of China(863 Program);Fundamental Research Funds for the Central Universities

摘要/Abstract

摘要：

随着重复数据删除次数的增加，系统中用于存储指纹索引的清单文件等元数据信息会不断累积，导致不可忽视的存储资源开销。因此，如何在不影响重复数据删除率的基础上，对重复数据删除过程中产生的元数据信息进行压缩，从而减小查重索引，是进一步提高重复数据删除效率和存储资源利用率的重要因素。针对查重元数据中存在大量冗余数据，提出了一种基于压缩近邻的查重元数据去冗算法Dedup²。该算法先利用聚类算法将查重元数据分为若干类，然后利用压缩近邻算法消除查重元数据中相似度较高的数据以获得查重子集，并在该查重子集上利用文件相似性对数据对象进行重复数据删除操作。实验结果表明，Dedup²可以在保持近似的重复数据删除比的基础上，将查重索引大小压缩50%以上。

关键词: 重复数据删除, 查重元数据, 近邻压缩规则

Abstract:

Building effective deduplication index in the memory could reduce disk access times and enhance chunk fingerprint lookup speed,which was a big challenge for deduplication algorithms in massive data environments.As deduplication data set had many samples with high similarity,a deduplication algorithm based on condensed nearest neighbor rule,which was called Dedup²,was proposed.Dedup²uses clustering algorithm to divide the original deduplication metadata into several categories.According to these categories,it employs condensed nearest neighbor rule to remove the highest similar data in the deduplication metadata.After that it can get the subset of deduplication metadata.Based on this subset,new data objects will be deduplicated based on the principle of data similarity.The results of experiments show that Dedup²can reduce the size of deduplication data set more than 50% effectively while maintain similar deduplication ratio.

Key words: deduplication, deduplication metadata, condensed nearest neighbor rule

姚文斌,叶鹏迪,李小勇,常静坤. 基于压缩近邻的查重元数据去冗算法设计[J]. 通信学报, 2015, 36(8): 1-7.

Wen-bin YAO,Peng-di YE,Xiao-yong LI,Jing-kun CHANG. Deduplication algorithm based on condensed nearest neighbor rule for deduplication metadata[J]. Journal on Communications, 2015, 36(8): 1-7.

图/表 6

图1

图2

图3

图4

图5

图6

参考文献 23

[1]	ZHU B , LI K , PATTERSON H . Avoiding the disk bottleneck in the data domain deduplication file system[A]. Proceedings of the 6th USENIX Conference on File and Storage Technologies,USENIX Association[C]. 2008. 1-14.
[2]	LILLIBRIDGE M , ESHGHI K , BHAGWAT D ,et al. Sparse indexing:large scale,inline deduplication using sampling and locality[A]. Proccedings of the 7th Conference on File and Storage Technologies,USENIX Association[C]. 2009. 111-123.
[3]	BHAGWAT D , ESHGHI K , LONG D ,et al. Extreme binning:scalable,parallel deduplication for chunk-based file backup[A]. In Modeling,Analysis ＆ Simulation of Computer and Telecommunication Systems,IEEE International Symposium[C]. IEEE, 2009. 1-9.
[4]	XIA W , JIANG H , FENG D ,et al. SiLo:a similarity-locality based near-exact deduplication scheme with low RAM overhead and high throughput[A]. Proceedings of the 2011 USENIX Annual Technical Conference(ATC),USENIX Association[C]. 2011. 26-28.
[5]	ARONOVICH L , ASHER R , BACHMAT E ,et al. The design of a similarity based deduplication system[A]. Proceedings of SYSTOR 2009,The Israeli Experimental Systems Conference[C]. ACM, 2009. 1-14.
[6]	ROMA?SKI B , HELDT ? ? , KILIAN W ,et al. Anchor-driven subchunk deduplication[A]. Proceedings of the 4th Annual International Conference on Systems and Storage[C]. 2011. 16-28.
[7]	ZHANG Z , BHAGWAT D , LITWIN W ,et al. Improved deduplication through parallel binning[A]. Performance Computing and Communications Conference(IPCCC),2012 IEEE 31st International[C]. 2012. 130-141.
[8]	DOUGLIS F , IYENGAR A . Application-specific deltaencoding via resemblance detection[A]. Proceedings of the 2003 USENIX Annual Technical Conference[C]. San Antonio,Texas, 2003. 113-126.
[9]	BRODER A Z , MITZENMACHER M . Network applications of Bloom filters:a survey[J]. Internet Mathematics, 2004,1(4): 485-509.
[10]	TAN L J , YAO W B , LIU Z Y.et al . CDFS:a cloud-based deduplication filesystem[J]. Advanced Science Letters,American Scientific Publishers, 2012,9(1): 855-860.
[11]	TEODOSIU D , BJORNER N , GUREVICH Y ,et al. Optimizing file replication over limited-bandwidth networks using remote differential compression[R]. Technical Report MSR-TR-2006-157,Microsoft Research, 2006.
[12]	YAO W B , YE P D . Simdedup:a new deduplication scheme based on simhash[A]. In Web-Age Information Management[C]. Springer Berlin Heidelberg, 2013. 79-88.
[13]	CHARIKAR M . Similarity estimation techniques from rounding algorithms[A]. Proc 34th Annual Symposium on Theory of Computing(STOC2002)[C]. 2002. 380-388.
[14]	MEISTER D , BRINKMANN A . Multi-level comparison of data deduplication in a backup scenario[A]. Proceedings of SYSTOR 2009,The Israeli Experimental Systems Conference[C].ACM, 2009.
[15]	MEYER D T , BOLOSKY W J . A study of practical deduplication[J]. ACM Transactions on Storage(TOS), 2012,7(4): 14.
[16]	WALLACE G , DOUGLIS F , QIAN H ,et al. Characteristics of backup workloads in production systems[A]. Proceedings of the Tenth USENIX Conference on File and Storage Technologies(FAST’12)[C]. 2012.
[17]	WEI J , JIANG H , ZHOU K ,et al. MAD2:a scalable high-throughput exact deduplication approach for network backup services[A]. Mass Storage Systems and Technologies(MSST),2010 IEEE 26th Symposium[C].IEEE, 2010. 1-14.
[18]	KAISER J , MEISTER D , BRINKMANN A ,et al. Design of an exact data deduplication cluster[A]. Mass Storage Systems and Technologies(MSST),2012 IEEE 28th Symposium[C].IEEE, 2012. 1-12.
[19]	BALACHANDRAN S , CONSTANTINESCU C . Sequence of hashes compression in data de-duplication[A]. Data Compression Conference,DCC 2008[C].IEEE, 2008.505.
[20]	CONSTANTINESCU C , PIEPER J , LI T . Block size optimization in deduplication systems[A]. Data Compression Conference,DCC'09[C].IEEE, 2009. 442-442.
[21]	ESHGHI K , LILLIBRIDGE M , WILCOCK L ,et al. Jumbo store:providing efficient incremental upload and versioning for a utility rendering service[A]. FAST[C]. 2007. 123-138.
[22]	MEISTER D , BRINKMANN A,Sü? T . File recipe compression in data deduplication systems[A]. Proceedings of 11th USENIX Conference on File and Storage Technologies(FAST)[C]. 2013. 175-182.
[23]	HART P E . The condensed nearest neighbor rule[J]. IEEE Transactions on Information Theory IT-14, 1968: 515-516.

基于压缩近邻的查重元数据去冗算法设计

Deduplication algorithm based on condensed nearest neighbor rule for deduplication metadata

在线阅读

PDF下载

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 6

参考文献 23

相关文章 2

Metrics

推荐阅读 0

[1]	熊金波,张媛媛,田有亮,应作斌,李琦,马蓉. 基于角色对称加密的云数据安全去重[J]. 通信学报, 2018, 39(5): 59-73.
[2]	熊金波,张媛媛,李凤华,李素萍,任君,姚志强. 云环境中数据安全去重研究进展[J]. 通信学报, 2016, 37(11): 169-180.