ZFS文件系统中双容错编码性能的研究  

Evaluating the Encoding Performance of the Double Fault-Tolerant Codes in ZFS

作　　者：张斌[1] 眭聚磊[1] 童健聪[1] 王刚[1] 刘晓光[1] 

机构地区：[1]南开大学信息技术科学学院,天津300071

出　　处：《计算机工程与科学》2010年第9期107-110,共4页Computer Engineering & Science

基　　金：国家863计划资助项目(2008AA01Z401);国家自然科学基金资助项目(60903028);教育部博士点基金课题(20070055054);天津市科技发展计划课题(08JCYBJC13000)

摘　　要：ZFS是Sun推出的一款革新性的文件系统,它支持用户构建镜像以及单容错/双容错的软Raid。其双容错编码方案采用的是Reed-Solomon编码(RS码)。由于RS码是基于有限域运算,编码/解码时间复杂性差是其根本性的缺陷。ZFS对写操作的处理采用的是聚合后追加的方式而非传统的覆盖方式,每次写操作都会进行一次编码计算。因此,编码计算性能是影响文件系统整体性能的重要因素之一。本文的工作是将RDP这一基于奇偶校验的双容错编码与ZFS相结合,替代Reed-Solomon编码,以优化文件系统写操作的性能。我们设计了Cache优化的RDP编码算法,在ZFS中进行了实现,并通过实验验证了这一方法的有效性。ZFS is an innovative file system invented by Sun,on which a user can construct mirror,single fault-tolerant or double fault-tolerant soft Raid systems. Its double-erasure code is the Reed-Solomon （RS） code. RS code based on calculating in the Galois field has a high time complexity. ZFS uses an appending writing strategy after the writing request aggregation instead of the traditional overwriting strategy. The performance of coding is one of the most important factors for ZFS,because coding calculation is necessary during every write operation. An approach to improving the writing performance by combining between ZFS and the RDP code whose computation is based on exclusive-or,instead of RS code,is shown. We have designed and implemented an algorithm by using cache＇s feature to improve the RDP coding performance in ZFS. Finally,the experiments have validated this approach.

关 键 词：ZFS 容错编码 RDP 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]