Failure Trends in a Large Disk Drive Population

By Eduardo Pinheiro, Wolf-Dietrich Weber, Luiz André Barroso. USENIX Conference on File and Storage Technologies, 2007

915 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/7827f3bbfe3e2713d69b74e1a87547f36eae3c58

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. Onur Mutlu
2. Jimmy J. Lin
3. Jeremie S. Kim
4. Donghyuk Lee
5. Yoongu Kim
6. Cheng Huang
7. Chris Fallin
8. Milo Polte

Citations in the corpus

• Flipping bits in memory without accessing them: An experimental study of DRAM disturbance errors (2014, background): “, more than 1% of hard-disk drives fail every year [54, 59]).”
• Privacy-Preserving Audit and Extraction of Digital Contents (2008, background): “Studies of deployed large-scale storage systems show that no storage service can be completely reliable; all have the potential to lose or corrupt customer data [4, 5, 19, 21].”
• An analysis of latent sector errors in disk drives (2007, methodology): “[12] analyzed SMART data associated with more than 100,000 disks taken over a nine month period.”
• Characterizing cloud computing hardware reliability (2010, unknown): “In doing so we build upon recen t large scale studies on hard disk [15, 17] and memory module fa ilure characteristics [18].”
• Disk Failures in the Real World: What Does an MTTF of 1, 000, 000 Hours Mean to You? (2007, unknown): “analyze disk replacement data from a large population of serial and parallel ATA drives [23].”
• An analysis of data corruption in the storage stack (2008, background): “While recent studies provide information on whole disk failures [Jiang et al. 2008; Pinheiro et al. 2007; Schroeder and Gibson 2007] and latent sector errors [Bairava­sundaram et al. 2007] that can aid system designers in handling these error conditions, very little is known about data…”
• Auditing to Keep Online Storage Services Honest (2007, background): “Unfortunately, this information is hard to obtain for storage services: Only a few studies describe failures and their causes in deployed large-scale storage systems [10, 12].”
• Understanding failures in petascale computers (2007, background): “More recently, several studies have been published on storage failures [32, 36, 25], all reporting ARR rates similar to the 3–6% range we observe in our data.”
• SSD Failures in Datacenters: What? When? and Why? (2016, methodology): “To quantify these failures, we use Annual-ized Failure Rate (AFR) as the metric, which is commonly used to report hardware failures [28, 29]: AFR = Total devices with failures Total device years in % For the latter, SSD speci-fications report AFRs be-tween 0.61% - 0.73%.”
• Data Security and Privacy in Cloud Computing (2014, unknown): “studied the error rate of hard-drives based on the historical data of hard-drive [41].”
• Book Reviews: Data-Intensive Text Processing with MapReduce by Jimmy Lin and Chris Dyer (2010, background): “In large clusters, disk failures are common [87] and RAM experiences more errors than one might expect [93].”
• Pyramid Codes: Flexible Schemes to Trade Space for Access Efficiency in Reliable Data Storage Systems (2007, result): “However, models that capture the failures of practical large scale systems are in general far more sophisticated, especially when failures could happen correlated, as recent observations suggest [ 19 ].”
• PLFS: a checkpoint filesystem for parallel applications (2009, background): “For large parallel simulations at LANL, and other High Performance Computing (HPC) sites, which can run for several months, restarting a calculation from the beginning is not viable, since the mean time to interrupt for today’s supercomputers is measured in days or hours and not months [31, 41].”