This article introduces the concept of abad square in aredundant array of inexpensive disks (RAID). Bad squares are used to prove upper limits on the reliability of the2d-parity arrangement when there is the possibility that astring of disks may fail simultaneously. Bad-square analysis motivates several optimal string layouts which achieve these limits. Bad squares also provide a means to calculate the mean time to data loss for a RAID layout, without the use of Monte Carlo simulation.
Key wordsRAID Disk array 2d-Parity Fault tolerance Disk strings Bad square
Unable to display preview. Download preview PDF.
- [BO]D. Blackston and C. Olson. Personal communication.Google Scholar
- [G]G. A. Gibson. Redundant disk arrays: Reliable, parallel secondary storage. Technical Report UCB/CSD 91/613, University of California, Berkeley, CA 94720, May 1991. Ph.D. thesis.Google Scholar
- [HGK+]L. Hellerstein, G. A. Gibson, R. M. Karp, R. H. Katz, and D. A. Patterson. Coding techniques for handling failures in large disk arrays.Algorithmica, this issue, pp. 182–208.Google Scholar
- [MK]J. Menon and J. Kasson. Methods for improved update performance of disk arrays, Technical Report RJ6928 (66034), IBM, November 1990.Google Scholar
- [PGK]D. A. Patterson, G. Gibson, and R. H. Katz. A case for redundant arrays of inexpensive disks (RAID).Proceedings of the 1988 ACM Conference on Management of Data (SIGMOD), Chicago, IL, June 1–3, 1988, pp. 109–116.Google Scholar
- [RB]A. L. N. Reddy and P. Banerjee. Graceful degradation of disk arrays.Proceedings of the Twenty-First International Symposium on Fault-Tolerant Computing, Montreal, June 25–27, 1991. IEEE Computer Society Press, New York.Google Scholar