Evaluating the impact of undetected disk errors in RAID systems

Eric W.D. Rozier; Wendy Belluomini; Veera Deenadhayalan; Jim Hafner; K. K. Rao; Pin Zhou

doi:10.1109/DSN.2009.5270353

Evaluating the impact of undetected disk errors in RAID systems

Eric W.D. Rozier, Wendy Belluomini, Veera Deenadhayalan, Jim Hafner, K. K. Rao, Pin Zhou

Electrical & Computer Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

19 Scopus citations

Abstract

Despite the reliability of modern disks, recent studies have made it clear that a new class of faults, UndetectedDisk Errors (UDEs) also known as silent data corruption events, become a real challenge as storage capacity scales. While RAID systems have proven effective in protecting data from traditional disk failures, silent data corruption events remain a significant problem unaddressed by RAID. We present a fault model for UDEs, and a hybrid framework for simulating UDEs in large-scale systems. The framework combines a multi-resolution discrete event simulator with numerical solvers. Our implementation enables us to model arbitrary storage systems and workloads and estimate the rate of undetected data corruptions. We present results for several systems and workloads, from gigascale to petascale. These results indicate that corruption from UDEs is a significant problem in the absence of protection schemes and that such schemes dramatically decrease the rate of undetected data corruption.

Original language	English (US)
Title of host publication	Proceedings of the 2009 IEEE/IFIP International Conference on Dependable Systems and Networks, DSN 2009
Pages	83-92
Number of pages	10
DOIs	https://doi.org/10.1109/DSN.2009.5270353
State	Published - Nov 26 2009
Event	2009 IEEE/IFIP International Conference on Dependable Systems and Networks, DSN 2009 - Lisbon, Portugal Duration: Jun 29 2009 → Jul 2 2009

Publication series

Name	Proceedings of the International Conference on Dependable Systems and Networks

Other

Other	2009 IEEE/IFIP International Conference on Dependable Systems and Networks, DSN 2009
Country	Portugal
City	Lisbon
Period	6/29/09 → 7/2/09

Keywords

Modeling
Silent data corruption
Simulation
Undetected disk errors

ASJC Scopus subject areas

Computer Networks and Communications
Hardware and Architecture
Software

Access to Document

10.1109/DSN.2009.5270353

Fingerprint Dive into the research topics of 'Evaluating the impact of undetected disk errors in RAID systems'. Together they form a unique fingerprint.

Cite this

Rozier, E. W. D., Belluomini, W., Deenadhayalan, V., Hafner, J., Rao, K. K., & Zhou, P. (2009). Evaluating the impact of undetected disk errors in RAID systems. In Proceedings of the 2009 IEEE/IFIP International Conference on Dependable Systems and Networks, DSN 2009 (pp. 83-92). [5270353] (Proceedings of the International Conference on Dependable Systems and Networks). https://doi.org/10.1109/DSN.2009.5270353

Evaluating the impact of undetected disk errors in RAID systems. / Rozier, Eric W.D.; Belluomini, Wendy; Deenadhayalan, Veera; Hafner, Jim; Rao, K. K.; Zhou, Pin.

Proceedings of the 2009 IEEE/IFIP International Conference on Dependable Systems and Networks, DSN 2009. 2009. p. 83-92 5270353 (Proceedings of the International Conference on Dependable Systems and Networks).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Rozier, EWD, Belluomini, W, Deenadhayalan, V, Hafner, J, Rao, KK & Zhou, P 2009, Evaluating the impact of undetected disk errors in RAID systems. in Proceedings of the 2009 IEEE/IFIP International Conference on Dependable Systems and Networks, DSN 2009., 5270353, Proceedings of the International Conference on Dependable Systems and Networks, pp. 83-92, 2009 IEEE/IFIP International Conference on Dependable Systems and Networks, DSN 2009, Lisbon, Portugal, 6/29/09. https://doi.org/10.1109/DSN.2009.5270353

Rozier EWD, Belluomini W, Deenadhayalan V, Hafner J, Rao KK, Zhou P. Evaluating the impact of undetected disk errors in RAID systems. In Proceedings of the 2009 IEEE/IFIP International Conference on Dependable Systems and Networks, DSN 2009. 2009. p. 83-92. 5270353. (Proceedings of the International Conference on Dependable Systems and Networks). https://doi.org/10.1109/DSN.2009.5270353

Rozier, Eric W.D. ; Belluomini, Wendy ; Deenadhayalan, Veera ; Hafner, Jim ; Rao, K. K. ; Zhou, Pin. / Evaluating the impact of undetected disk errors in RAID systems. Proceedings of the 2009 IEEE/IFIP International Conference on Dependable Systems and Networks, DSN 2009. 2009. pp. 83-92 (Proceedings of the International Conference on Dependable Systems and Networks).

@inproceedings{78124372c129433c9cd80db322523523,

title = "Evaluating the impact of undetected disk errors in RAID systems",

abstract = "Despite the reliability of modern disks, recent studies have made it clear that a new class of faults, UndetectedDisk Errors (UDEs) also known as silent data corruption events, become a real challenge as storage capacity scales. While RAID systems have proven effective in protecting data from traditional disk failures, silent data corruption events remain a significant problem unaddressed by RAID. We present a fault model for UDEs, and a hybrid framework for simulating UDEs in large-scale systems. The framework combines a multi-resolution discrete event simulator with numerical solvers. Our implementation enables us to model arbitrary storage systems and workloads and estimate the rate of undetected data corruptions. We present results for several systems and workloads, from gigascale to petascale. These results indicate that corruption from UDEs is a significant problem in the absence of protection schemes and that such schemes dramatically decrease the rate of undetected data corruption.",

keywords = "Modeling, Silent data corruption, Simulation, Undetected disk errors",

author = "Rozier, {Eric W.D.} and Wendy Belluomini and Veera Deenadhayalan and Jim Hafner and Rao, {K. K.} and Pin Zhou",

year = "2009",

month = nov,

day = "26",

doi = "10.1109/DSN.2009.5270353",

language = "English (US)",

isbn = "9781424444212",

series = "Proceedings of the International Conference on Dependable Systems and Networks",

pages = "83--92",

booktitle = "Proceedings of the 2009 IEEE/IFIP International Conference on Dependable Systems and Networks, DSN 2009",

note = "2009 IEEE/IFIP International Conference on Dependable Systems and Networks, DSN 2009 ; Conference date: 29-06-2009 Through 02-07-2009",

}

TY - GEN

T1 - Evaluating the impact of undetected disk errors in RAID systems

AU - Rozier, Eric W.D.

AU - Belluomini, Wendy

AU - Deenadhayalan, Veera

AU - Hafner, Jim

AU - Rao, K. K.

AU - Zhou, Pin

PY - 2009/11/26

Y1 - 2009/11/26

N2 - Despite the reliability of modern disks, recent studies have made it clear that a new class of faults, UndetectedDisk Errors (UDEs) also known as silent data corruption events, become a real challenge as storage capacity scales. While RAID systems have proven effective in protecting data from traditional disk failures, silent data corruption events remain a significant problem unaddressed by RAID. We present a fault model for UDEs, and a hybrid framework for simulating UDEs in large-scale systems. The framework combines a multi-resolution discrete event simulator with numerical solvers. Our implementation enables us to model arbitrary storage systems and workloads and estimate the rate of undetected data corruptions. We present results for several systems and workloads, from gigascale to petascale. These results indicate that corruption from UDEs is a significant problem in the absence of protection schemes and that such schemes dramatically decrease the rate of undetected data corruption.

AB - Despite the reliability of modern disks, recent studies have made it clear that a new class of faults, UndetectedDisk Errors (UDEs) also known as silent data corruption events, become a real challenge as storage capacity scales. While RAID systems have proven effective in protecting data from traditional disk failures, silent data corruption events remain a significant problem unaddressed by RAID. We present a fault model for UDEs, and a hybrid framework for simulating UDEs in large-scale systems. The framework combines a multi-resolution discrete event simulator with numerical solvers. Our implementation enables us to model arbitrary storage systems and workloads and estimate the rate of undetected data corruptions. We present results for several systems and workloads, from gigascale to petascale. These results indicate that corruption from UDEs is a significant problem in the absence of protection schemes and that such schemes dramatically decrease the rate of undetected data corruption.

KW - Modeling

KW - Silent data corruption

KW - Simulation

KW - Undetected disk errors

UR - http://www.scopus.com/inward/record.url?scp=70450198064&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70450198064&partnerID=8YFLogxK

U2 - 10.1109/DSN.2009.5270353

DO - 10.1109/DSN.2009.5270353

M3 - Conference contribution

AN - SCOPUS:70450198064

SN - 9781424444212

T3 - Proceedings of the International Conference on Dependable Systems and Networks

SP - 83

EP - 92

BT - Proceedings of the 2009 IEEE/IFIP International Conference on Dependable Systems and Networks, DSN 2009

T2 - 2009 IEEE/IFIP International Conference on Dependable Systems and Networks, DSN 2009

Y2 - 29 June 2009 through 2 July 2009

ER -