Design and modeling of clustered RAID

Author

Merchant, A. ; Yu, P.S.

Author_Institution

NEC USA, Princeton, NJ, USA

fYear

1992

fDate

8-10 July 1992

Firstpage

140

Lastpage

149

Abstract

A redundant array of independent disks (RAID) consisting of G disks provides protection against single disk failures by adding one parity block for each G-1 data blocks. In a clustered RAID, the G data/parity blocks are distributed over a cluster of C disks, C>G, thus reducing the additional load on each disk due to a single disk failure. The authors present a fast algorithm for distributing parity groups of size G over C disks. They create mappings for each parity group based on almost-random permutations. An analytical model is constructed to predict recovery time and read/write performance. The analysis shows that the clustered RAID is significantly more tolerant of disk failure than the basic RAID scheme. Both recovery time and performance degradation during recovery are substantially reduced in the clustered RAID.<>

Keywords

fault tolerant computing; magnetic disc storage; G-1 data blocks; almost-random permutations; analytical model; clustered RAID; data/parity blocks; disk failure; mappings; modeling; performance degradation; read/write performance; recovery time; redundant array of independent disks; single disk failures; Costs; Degradation; Failure analysis; Laboratories; Mirrors; National electric code; Performance gain; Predictive models; Protection; Throughput;

fLanguage

English

Publisher

ieee

Conference_Titel

Fault-Tolerant Computing, 1992. FTCS-22. Digest of Papers., Twenty-Second International Symposium on

Conference_Location

Boston, MA, USA

Print_ISBN

0-8186-2875-8

Type

conf

DOI

10.1109/FTCS.1992.243606

Filename

243606