A Highly Accurate Method for Assessing Reliability of Redundant Arrays of Inexpensive Disks (RAID)

Elerath, J.G.; Pecht, Michael

doi:10.1109/tc.2008.163

Cited by 55 publications

(44 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This new model is based on statistical principles associated with nonHomogeneous Poisson processes, evaluated using Monte Carlo simulation. The simulation results follow field data closely [2,3,4], but this improved accuracy comes at the expense of usability, since simulations require specialized code and extensive computation.…”

Section: Introductionmentioning

confidence: 80%

See 1 more Smart Citation

A simple equation for estimating reliability of an N+1 redundant array of independent disks (RAID)

Elerath

2009

2009 IEEE/IFIP International Conference on Dependable Systems &Amp; Networks

View full text Add to dashboard Cite

This paper develops an equation that provides a good approximation of the expected number of doubledisk failures (data losses) in an (N+1) redundant array of inexpensive disks (RAID) as a function of time. This paper includes the statistical bases for the equation, sources of error and inaccuracies due to approximations, and limitations of its use. The equation is simple and can be evaluated using a hand held calculator or basic spreadsheet. Accuracy depends on four input distributions, which include operational failures, operational failure restorations, latent defects, and data scrubbing. Failure and restoration distributions may represent nonhomogeneous Poisson processes and, therefore, not have constant rates. Results of the equations are compared to two other hand calculation methods, to a highly accurate Monte Carlo simulation, and to actual field data for more than 10,000 RAID groups composed of 14 drives.

show abstract

Section: Introductionmentioning

confidence: 80%

“…Often times, the mean is converted into a more concrete term, the number of double disk failures (DDFs) per year. Unfortunately, the derived number of DDFs per year as a function of time based on MTTDL does not agree with actual field data [2].…”

Section: Introductionmentioning

confidence: 82%

A simple equation for estimating reliability of an N+1 redundant array of independent disks (RAID)

Elerath

2009

2009 IEEE/IFIP International Conference on Dependable Systems &Amp; Networks

View full text Add to dashboard Cite

show abstract

“…A previous study [5] showed that the MTTDL errors are so large that the equation is not even a reasonable approximation. This study shows that even when latent defects are included, RAID group failure frequency depends greatly on the Ld scrub times set by the operating system priorities.…”

Section: Figure 4 -Effects Of Recovery Distributions Group Size = 23mentioning

confidence: 98%

“…HDDs generally do not have constant failure rates [5], and a 2-parameter Weibull distribution has been shown to be a reasonable distributional fit for operational failures. Latent defects are often caused by contamination of other causes inherent to the HDD design [3].…”

Section: Operational Failures and Latent Defectsmentioning

confidence: 99%

RAID-6 system reliability dependence on recovery, disk scrubbing, and group size

Elerath

2016

2016 Annual Reliability and Maintainability Symposium (RAMS)

View full text Add to dashboard Cite

Analyses of 108 combinations of hard disk drive failure, restoration, latent defect and scrubbing distributions were analyzed. A subset of 8 combinations for 2 different RAID group sizes are presented to illustrate the impact that RAID group size, recovery times, and disk scrubbing of latent defect has on the occurrence of system (RAID group) failures for 10 years. Increasing the group size or extending the recovery or scrub times increases the expected number of failures, but the increase is not predictable based on the input distributions. RAID group data loss events show the same time-dependence, increasing or decreasing, as the HDD operational failures.

show abstract

“…Mean time to data loss (MTTDL) is a reliability index of storage systems that can easily be calculated and has been widely used in the field [7]. However, some authors report discrepancies between the actual rate of data loss and MTTDL [7], [8].…”

Section: Related Workmentioning

confidence: 99%

Reliability and Failure Impact Analysis of Distributed Storage Systems with Dynamic Refuging

Akutsu

Ueda

Chiba

et al. 2016

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARYIn recent data centers, large-scale storage systems storing big data comprise thousands of large-capacity drives. Our goal is to establish a method for building highly reliable storage systems using more than a thousand low-cost large-capacity drives. Some large-scale storage systems protect data by erasure coding to prevent data loss. As the redundancy level of erasure coding is increased, the probability of data loss will decrease, but the increase in normal data write operation and additional storage for coding will be incurred. We therefore need to achieve high reliability at the lowest possible redundancy level. There are two concerns regarding reliability in large-scale storage systems: (i) as the number of drives increases, systems are more subject to multiple drive failures and (ii) distributing stripes among many drives can speed up the rebuild time but increase the risk of data loss due to multiple drive failures. If data loss occurs by multiple drive failure, it affects many users using a storage system. These concerns were not addressed in prior quantitative reliability studies based on realistic settings. In this work, we analyze the reliability of large-scale storage systems with distributed stripes, focusing on an effective rebuild method which we call Dynamic Refuging. Dynamic Refuging rebuilds failed blocks from those with the lowest redundancy and strategically selects blocks to read for repairing lost data. We modeled the dynamic change of amount of storage at each redundancy level caused by multiple drive failures, and performed reliability analysis with Monte Carlo simulation using realistic drive failure characteristics. We showed a failure impact model and a method for localizing the failure. When stripes with redundancy level 3 were sufficiently distributed and rebuilt by Dynamic Refuging, the proposed technique turned out to scale well, and the probability of data loss decreased by two orders of magnitude for systems with a thousand drives compared to normal RAID. The appropriate setting of a stripe distribution level could localize the failure.

show abstract

A Highly Accurate Method for Assessing Reliability of Redundant Arrays of Inexpensive Disks (RAID)

Cited by 55 publications

References 17 publications

A simple equation for estimating reliability of an N+1 redundant array of independent disks (RAID)

A simple equation for estimating reliability of an N+1 redundant array of independent disks (RAID)

RAID-6 system reliability dependence on recovery, disk scrubbing, and group size

Reliability and Failure Impact Analysis of Distributed Storage Systems with Dynamic Refuging

Contact Info

Product

Resources

About