Reliability Analysis of Energy-Efficient Parallel Storage Systems

作者: Shu Yin , Kenli Li , Xiao Qin , Xiaojun Ruan , Haiquan Chen

DOI: 10.1109/HPCC.AND.EUC.2013.205

关键词: LocalityReliability (statistics)Energy storageCover (telecommunications)RAIDComputer scienceParallel computingReliability engineeringEfficient energy useData recovery

摘要: Existing reliability models evaluate the lifetime of storage systems well. However, few aim to analysis parallel systems, especially equipped with energy-efficient solutions. MREED model provides an inspiring idea for RAIDs. Unfortunately, has its disadvantage that it can RAID-0 only, in which a basic method data-stripping is applied. In this paper, we extend cover most RAID setups including RAID-1, introduces disk-mirroring techniques, and RAID-5, applies data-parity techniques based on schemes data recovery purpose. The then applied existing well-known power-aware system-PARAID. Experimental results indicate locality affection parity impaction PARAID. Furthermore, investigate fundamental trade-off between energy-efficiency terms energy-saving systems. A tradeoff curve provided justify whether worth sacrifice better energy efficiency.

参考文章(31)
Fred Douglis, Brian Marsh, P. Krishnan, Thwarting the power-hungry disk usenix winter technical conference. pp. 23- 23 ,(1994)
David P. Helmbold, Darrell D.E. Long, Tracey L. Sconyers, Bruce Sherrod, Adaptive disk spin—down for mobile computers Mobile Networks and Applications. ,vol. 5, pp. 285- 297 ,(2000) , 10.1023/A:1019129116852
Taliver Heath, Enrique V. Carrera, Ricardo Bianchini, Eduardo Pinheiro, Load balancing and unbalancing for power and performance in cluster-based systems ,(2001) , 10.7282/T3-AGFW-YT73
Paul Horton, Thomas Anderson, Kester Li, Roger Kumpf, A Quantitative analysis of disk drive power management in portable computers usenix winter technical conference. pp. 22- 22 ,(1994)
Michael Scott, Athanasios Papathanasiou, Power-efficient Server-class Performance from Arrays of Laptop Disks ,(2004)
J.G. Elerath, Specifying reliability in the disk drive industry: No more MTBF's reliability and maintainability symposium. pp. 194- 199 ,(2000) , 10.1109/RAMS.2000.816306
Sudhanva Gurumurthi, Anand Sivasubramaniam, Mahmut Kandemir, Hubertus Franke, DRPM Proceedings of the 30th annual international symposium on Computer architecture - ISCA '03. ,vol. 31, pp. 169- 181 ,(2003) , 10.1145/859618.859638
Sankaran Sivathanu, Ling Liu, Cristian Ungureanu, Modeling the performance and energy of storage arrays international conference on green computing. pp. 229- 242 ,(2010) , 10.1109/GREENCOMP.2010.5598308
Shudong Jin, Azer Bestavros, G ISMO ACM SIGMETRICS Performance Evaluation Review. ,vol. 29, pp. 2- 10 ,(2001) , 10.1145/507553.507554
Andreas Weissel, Björn Beutel, Frank Bellosa, Cooperative I/O ACM SIGOPS Operating Systems Review. ,vol. 36, pp. 117- 129 ,(2002) , 10.1145/844128.844140