摘要
保证数据存储可靠性是信息领域面临的一个重要问题,而研究保证数据存储可靠性的算法则是其中的核心技术和热点问题。从容错度、编解码性能等多方面考虑提出了很多算法。按照存储性能和编码方式对目前的数据存储可靠性算法进行分类;对国际上提出的典型算法进行了分析,包括RS、CRS、EVENODD、STAR、X-Code、WEAVER、HoVer和Tornado算法等,讨论了它们的工作原理,提出了一套对相关算法进行性能分析的性能评价指标。总结了数据存储可靠性算法具有的特点和需要进一步研究的问题。
It is challenging to ensure data reliable storage in information research community, among which the research of data reliable storage algorithms is one of the key issues. Considering fault tolerance, encoding and decoding performance etc, many algorithms are proposed. In this paper, data reliable storage algorithms were classified according to their storage performance and coding manner. Some typical algorithms, such as RS, CRS, EVENODD, STAR, X-Code, WEAVER, HoVer and Tornado code were analyzed. The coding principles of these algorithms were discussed. A set of performance criterions which can be used for evaluating storage performance were presented. Finally the characteristic of good algorithms and research directions in the future were summarized.
出处
《解放军理工大学学报（自然科学版）》
EI
2007年第3期239-245,共7页
Journal of PLA University of Science and Technology(Natural Science Edition)
基金
国家自然科学基金资助项目(60403043)
关键词
磁盘阵列
最大距离可分的
奇偶校验矩阵编码算法
低密度校验码算法
RAID（redundant array of independent disk）
MDS（maximun distance separable）
parity array codes
low-density parity codes
