Uniform parity group distribution in disk arrays with multiple failures

Ng, Spencer W.; Mattson, Robert L. R.

doi:10.1109/12.278490

Cited by 16 publications

(10 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Balanced Incomplete Block Designs (BIBD) [16,7] and Nearly Random Permutations (NRP) [19] are two implementations of CRAID to balance the disk update load [33]. BIBD has the shortcoming that it is not available for all values of N and α [33].…”

Section: Methods To Improve Rebuild Performancementioning

confidence: 99%

Rebuild processing in RAID5 with emphasis on the supplementary parity augmentation method[37]

Thomasian¹

2012

SIGARCH Comput. Archit. News

View full text Add to dashboard Cite

The rotated parity RAID5 disk array tolerates single disk failures by continuing operation by on-demand reconstruction of data blocks of the failed disk, until the systematic reconstruction of the contents of the failed disk is completed by the rebuild process on a spare disk. Supplementary Parity Augmentation (SPA), unlike the pyramid code, which has two parities covering half of the arrays disks each, extends RAID5's P parity with an additional S parity, which covers half of the disks. The extra load with respect to RAID5 of updating the S parity by one half of the disks is compensated by the more efficient on demand reconstructtion and rebuild processing when a disk fails. Although SPA has the same disk space redundancy level as RAID6, unlike RAID6 it can only deal with roughly half of all possible double disk failure cases for eight disks. For rebuild processing SPA reads half of the disks required by RAID5 and this leads to a higher Mean Time to Data Loss than RAID5, since fewer Latent Sector Errors are encountered. We review performance and reliability modeling of RAID5 arrays to provide insights into SPA's performance and reliability, which cannot be gained from numerical results alone. SPA is outperformed by the Intra-Disk Redundancy schemes combined with RAID5, which results in RAID6's reliability and RAID5 performance.,

show abstract

Section: Methods To Improve Rebuild Performancementioning

confidence: 99%

Rebuild processing in RAID5 with emphasis on the supplementary parity augmentation method[37]

Thomasian¹

2012

SIGARCH Comput. Archit. News

View full text Add to dashboard Cite

show abstract

“…The Balanced Incomplete Block Design (BIBD) method to implement parity declustering is described in [32,16], while the Nearly Random Permutation (NRP) method is described in [29]. Both methods are also described in [51].…”

Section: Alternatives To Updating Check Blocksmentioning

confidence: 99%

Why specialized disks for composite operations may be unnecessary

Thomasian

2010

SIGARCH Comput. Archit. News

View full text Add to dashboard Cite

Disk arrays with erasure coding such as RAID5 and RAID6 incur four and six disk accesses respectively for updating data and check blocks. The small write penalty can be reduced by the Read-Modify-Write (RMW) composite operations to update data and associated check blocks. The Disk Architecture with Composite Operation (DACO) is a proposal to eliminate the disk rotation associated with RMWs, by using a complex read/write head, which allows the writing of a block immediately after reading and modifying it without needing an extra disk rotation. We argue that the extra cost associated with DACO may not be justifiable, because it is not expected to have a significant impact on RAID performance. Furthermore an XOR capability is still required at the disk array controller for reconstructing missing data blocks. A duplexed Nonvolatile Storage (NVS) cache at the disk array controller provides the same reliability as magnetic disks and allows fast writes, i.e., writing to disk is considered completed as soon as data is written onto NVS. Deferring the destaging of data blocks from NVS allows these blocks to be be overwritten, obviating unnecessary disk writes. This also allows neighboring dirty blocks to be destaged in batches, so that a higher disk access efficiency is attained. Disks with multiple arms can also be used to make the processing of RMW requests more efficient, while disks with multiple R/W heads on one arm have little effect on RMW requests. In addition there are alternative methods to update check blocks, such as floating parities, parity logging, the reconstruct write method, log structured arrays, and variable scope parity protection.

show abstract

“…Given N and G there are N G permutations, but this number may be too large to yield a balanced allocation on finite capacity disks, at least for the fraction of disk capacity that is utilized [6]. BIBDs were proposed in [7,8] as an alternative. NRP is an alternative to BIBD to balance the load for updating parities [9].…”

Section: Appendix A: Clustered Raid Designsmentioning

confidence: 99%

“…In Appendix A we describe three clustered RAID layouts. Balanced incomplete block designs (BIBDs) [7,8] and nearly random permutations (NRPs) [9] are applicable to RAID5 and RAID6. RM2, based on the redundancy matrix, is a specialized array designed for tolerating two disk failures [10] which also happens to be a clustered RAID, since its PG size is smaller than the number of disks.…”

mentioning

confidence: 99%

Clustered RAID Arrays and Their Access Costs

Thomasian

2005

The Computer Journal

View full text Add to dashboard Cite

RAID5 (resp. RAID6) are two popular RAID designs, which can tolerate one (resp. two) disk failures, but the load of surviving disks doubles (resp. triples) when failures occur. Clustered RAID5 (resp. RAID6) disk arrays utilize a parity group size G, which is smaller than the number of disks N, so that the redundancy level is 1/G (resp. 2/G). This enables the array to sustain a peak throughput closer to normal mode operation; e.g. the load increase for RAID5 in processing read requests is given by α = (G − 1)/(N − 1). Three methods to realize clustered RAID are balanced incomplete blocks designs and nearly random permutations, which are applicable to RAID5 and RAID6, and RM2 where each data block is protected by two parity disks. We derive cost functions for the processing requirements of clustered RAID in normal and degraded modes of operation. For given disk characteristics, the cost functions can be translated into disk service times, which can be used for the performance analysis of disk arrays. Numerical results are used to quantify the level of load increase in order to determine the value of G which maintains an acceptable level of performance in degraded mode operation.

show abstract

Uniform parity group distribution in disk arrays with multiple failures

Cited by 16 publications

References 5 publications

Rebuild processing in RAID5 with emphasis on the supplementary parity augmentation method[37]

Rebuild processing in RAID5 with emphasis on the supplementary parity augmentation method[37]

Why specialized disks for composite operations may be unnecessary

Clustered RAID Arrays and Their Access Costs

Contact Info

Product

Resources

About