Abstract
Variation of RAID with emphasis on RAID(4+k),1⩽k⩽3, which use k check disks to protect against as many disk failures. Operation in normal, degraded, and rebuild modes and their performance analysis. Clustered RAID5 uses parity group size G smaller than the number of disks to reduce disk overload when a disk fails. Balanced Incomplete Block Designs - BIBD and Nearly Random Permutations - NRP are two methods to place parity blocks to balance disk loads. Data layouts are used to reduce disk load increase when a disk fails. Various methods for rebuild processing including the Vacationing Server Model - VSM and Permanent Customer Model - PCM are presented. Analysis verified by simulation showed that VSM outperforms PCM in rebuild time and mean response times of external requests. IntraDisk redundancy - IDR and disk scrubbing methods to deal with Latent Sector Errors - LSEs, which may lead to unsuccessful rebuild are presented. RAID6 utilizes Reed-Solomon - RS codes, but parities incurring lower computational cost than RS can be used to implement Maximum Distance Separable - MDS 2-Disk-Failure-Tolerant - 2DFT arrays, such as EVENODD, RDP, and X-code. Taking advantage of two check codes can be applied to rebuild a single disk in the case of X-code, but this method is not applicable to EVENODD and RDP which have small symbols for efficient access. Coding technique applicable to distributed storage systems is also discussed.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
