首页 | 本学科首页   官方微博 | 高级检索  
     

多级存储数据价值评定研究
引用本文:张菁,任怡,廖湘科.多级存储数据价值评定研究[J].计算机研究与发展,2012(Z1):336-340.
作者姓名:张菁  任怡  廖湘科
作者单位:国防科学技术大学计算机学院
基金项目:国家“核高基”科技重大专项课题基金项目(2010ZX01045-001-002-5,2010ZX01045-001-007-2,2010ZX01036,2010ZX01037)
摘    要:数据价值评定是多级存储中必不可少的重要部分,评值方法是否准确和合适直接影响了整个多级存储系统的优劣.但评定数据价值时需考虑的因素很多,如何选择和综合权衡这些因素是一个难点.另外,目前还没有评价数据价值评定方法优劣的权威标准.本文对多级存储中的数据价值评定问题进行了研究,提出通过准确度、适应度和依赖度3方面指标评价数据价值评定方法的优劣;对比了当前的各种方法并归纳其发展趋势;展望了未来研究方向,指出不同优化目标下的数据价值评定和结合固态硬盘的多级存储数据价值评定在未来的研究价值.

关 键 词:数据价值评定  多级存储  基于策略  数据访问形态  固态硬盘

Data Valuation on Multi-tiered Storage
Zhang Jing, Ren Yi, and Liao Xiangke.Data Valuation on Multi-tiered Storage[J].Journal of Computer Research and Development,2012(Z1):336-340.
Authors:Zhang Jing  Ren Yi  and Liao Xiangke
Affiliation:(College of Computer, National University of Defense Technology, Changsha 410073)
Abstract:Data valuation is very important for multi-tiered storage. The accuracy and the adaptivity of data valuation methods are the key issues for overall effectiveness of multi-tiered storage system. There are large amounts of factors which may determine the value of data, so how to choose factors and synthetically take them into account is a challenge. So far, there is no authoritative standard for effectively evaluating various data valuation methods. In this paper, we propose an evaluation measurement for different data valuation methods based on three dimensions, i.e., accuracy, adaptivity and dependentivity. Then we compare and evaluate the existing data valuation methods by the proposed measurement. Finally, we discuss future challenges and directions. We believe that data valuation for diverse optimized targets and for SSD-based multi-tiered storage will be two major directions.
Keywords:data valuation  multi-tiered storage  policy-based  data access pattern  solid-state drives
本文献已被 CNKI 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号