基于多属性度量的数据分级访问模型研究

计算机科学 ›› 2013, Vol. 40 ›› Issue (Z11): 165-169.

基于多属性度量的数据分级访问模型研究

施光源,张宇

浪潮集团有限公司济南250101;浪潮集团有限公司济南250101

出版日期:2018-11-16 发布日期:2018-11-16
基金资助:
本文受国家“九七三”重点基础研究计划基金项目(2012CB724101)资助

Hierarchical Storage Access Model Based on Multi-Attributes Measurement

SHI Guang-yuan and ZHANG Yu

Online:2018-11-16 Published:2018-11-16

摘要/Abstract

摘要： 随着云计算的迅速发展,云存储已成为企业关键信息服务的重要途径。但是,受限于存储资源性能以及大数据等影响,用户往往需要忍受较长的访问延时。为了缓解这种情况,人们提出了智能数据管理技术,用于有效管理大量数据以及降低用户的访问延迟,提高云计算的服务质量。提出一种基于多属性分析的存储端数据分级访问模型。模型通过对被管理数据对象的动态、静态属性进行统计分析来提取属性中的关键信息,并依此进行数据管理决策,将冷/热数据迁移至对应层级,以便能够在合理规划存储资源的同时提高存储系统的访问性能。性能测试实验的结果表明,该模型具有较好的整体性能。

关键词: 智能数据管理,存储访问模型,分级存储,数据迁移

Abstract: With the rapid development of cloud computing, cloud storage has become an important way of business-critical information services. However, limited by storage resource performance and impact data, users often need to endure a long delay of access. In order to alleviate the situation, intelligent data management technologies have been proposed, to effectively manage large amounts of data, as well as reduce access latency, improving the quality of cloud computing services. Proposing an Hierarchical Storage Access Model based on Multi-Attributes Measurement. Model managed through the statistical analysis of the static and dynamic properties of the data object, extracting key information from the properties, data management and in accordance with this decision, cold/hot data migration to the corresponding levels so that you can reasonable while planning your storage resources to improve storage system performance. Performance test results of the experiment showed that the model has good overall performance.

Key words: Intelligent data magement,Storage access model,Hierarchical storage,Data migration

施光源,张宇. 基于多属性度量的数据分级访问模型研究[J]. 计算机科学, 2013, 40(Z11): 165-169. https://doi.org/

SHI Guang-yuan and ZHANG Yu. Hierarchical Storage Access Model Based on Multi-Attributes Measurement[J]. Computer Science, 2013, 40(Z11): 165-169. https://doi.org/

参考文献

[1] Massiglia P.Exploiting Multi-tier File Storage Effectively 2009[R/OL].http//www.snia.org/educa tion/tutorials/2009/spring/file/PaulMassiglia_Expoiting_Multi-Tier_File_StorageV05.pdf
[2] 赵晓南,李战怀,曾雷杰,等.分级存储管理技术研究[J].计算机研究与发展,2001,48(Suppl.):105-111
[3] Tim L C,Simth C H M, Miller E L.An improved long-term file-usage prediction algorithm.1999.http://users.soe.ucsc.edu/~elm/Papers/cmg99.pdf
[4] Mesnier M, Thereska E, Ganger G R, et al.File classification in self-* storage systems[C]∥Proc of the 1^st Int Conf on Autonomic Computing.Washington, D C:IEEE,2004:44-51
[5] Jin Hai, Xiong Muzhou, Wu Song.Information values evaluation model for ILM[C]∥Proc of 9^th ACIS Int Conf on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD2008).Washington, D C:IEEE, 2008:543-548
[6] Shah G,Voruganti K, Shivam P, et al.ACE:Classification for information lifecycle management[C]∥Proc of NASA Mass Storage Systems and Technologies (MSST 2006).Washington, D C:IEEE, 2006
[7] Gong Zhang, Lawre nce Chiu,Clem Dickey, et al.Automated Look ahead Data Migration SSD-enabled Multi-tiered S torage Systems[C]∥ Proc of 26th IEEE(MSS T2010) Symposium on Massive Storage Systems.Incline Village, Nevada,2010
[8] 聂雪军,秦磊华,周敬利,等.内容感知存储系统中自动分级存储模型研究[J].小型微型计算机系统,2011,32(6)
[9] Guerra J, Pucha H, Glider J, et al.Cost Effective Storage using Extent Based Dynamic Tiering[C]∥The Proceeding of USENIX Conference on File and Storage Technologiesp.2011:273-286
[10] 宋丽娜,戴华东,任怡.基于海量数据存储系统多级存储介质的热点数据区分方法[J].计算机研究与发展,2012,49(Suppl.):6-11

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed