Abstract
With the application of MLC (multi-level cell) and TLC (triple-level cell) techniques, the price of NAND flash memory based SSD (solid-state drive) decreases rapidly with increasing capacity. However, these techniques negatively influence the reliability of SSD as they lead to a larger number of raw flash memory errors. When multiple such reliability degraded SSDs organized in a RAID structure SSD failures could occur. Thus, a reliability-aware data reconstruction mechanism that can quickly recover the data of a failed SSD onto a replacement SSD becomes essential. In this paper, we propose a reliability-driven data reconstruction strategy called PDB (Popular Data Backup) for RAID4 and SPD-RAID4 (Splitting Parity Disk - RAID4), a variant of RAID4. PDB collaboratively backups popular data among data SSDs to achieve a shorter “window of vulnerability”. Experimental results demonstrate that compared with the traditional SOR (Stripe Oriented Reconstruction) method PDB can shorten reconstruction time up to 31.3%.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
Agrawal, N., Prabhakaran, V., Wobber, T., Davis, J.D., Manasse, M., Panigrahy, R.: Design Tradeoffs for SSD Performance. In: USENIX Ann. Technical Conference, pp. 57–70. USENIX Association, Berkeley (2008)
Cai, Y., Haratsch, E.F., Mutlu, O., Mai, K.: Threshold Voltage Distribution in MLC NAND Flash Memory: Characteristization, Analysis, and Modeling. In: The Conf. on Design, Automation and Test in Europe, EDA Consortium San Jose, pp. 1285–1290 (2013)
Xie, T., Sharma, A.: Collaboration-Oriented Data Recovery for Mobile Disk Arrays. In: 29th Int’l Conf. on Distributed Computed Systems, Montreal, pp. 631–638 (June 2009)
Patterson, D.A., Gibson, G., Katz, R.H.: A Case for Redundant Arrays for Inexpensive Disks (RAID). In: Boral, H., Larson, P.A. (eds.) 1988 ACM SIGMOD Int’l Conf. on Management of Data, pp. 109–116. ACM, New York (1988)
Im, S., Shin, D.: Flash-Aware RAID Techniques for Dependable and High-Performance Flash Memory SSD. J. IEEE Transaction on Computer. 6(1), 80–92 (2011)
Lee, Y., Jung, S., Song, Y.H.: FRA: A Flash-aware Redundant Array of Flash Storage Devices. In: 7th IEEE/ACM Int’l Conf. on Hardware/Software Codesign and System Synthesis, pp. 163–172. ACM, New York (2009)
Wu, S.Z., Jiang, H., Feng, D., Tian, L., Mao, B.: Workout: I/O Workload Outsourcing for Boosting RAID Reconstruction Performance. In: 7th USENIX Conf. on FAST, pp. 239–252. USENIX Association, Berkeley (2009)
Tian, L., Feng, D., Jiang, H., Zhou, K., Zeng, L.F., Chen, J.X., Wang, Z.K., Song, Z.L.: PRO: A Popularity-Based Multi-Threaded Reconstruction Optimization for RAID-Structured Storage Systems. In: 5th USENIX Conf. on FAST, pp. 277–290. USENIX Association, Berkeley (2007)
Xie, T., Wang, H.: MICRO: A Multilevel Caching-Based Reconstruction Optimization for Mobile Storage Systems. J. IEEE Transactions on Computers 57(10), 1386–1398 (2008)
Pan, W., Liu, F., Xie, T., Gao, Y.Y., Ouyang, Y.M., Chen, T.: SPD-RAID4: Splitting Parity Disk for RAID4 Structured Parallel SSD Arrays. In: 15th Int’l Conf. on High Performance Computing and Communications. IEEE Press, Zhangjiajie (2013)
Hu, Y., Jiang, H., Feng, D., Tian, L., Luo, H., Zhang, S.P.: Performance Impact and Interplay of SSD Parallelism through Advanced Commands, Allocation Strategy and Data Granularity. In: Int’l Conf. on Supercomputing, pp. 96–107. ACM, New York (2011)
Holland, M.: Online Data Reconstruction in Redundant Disk Arrays. In: PhD Dissertation CMU-CS-94-164, Carnegie Mellon Univ., Pittsburgh (1994)
Holland, M., Gibson, G.A., Siewiorek, D.P.: Fast, On-Line Failure Recovery in Redundant Disk Arrays. In: 23rd Ann. Int’l Symp. on Fault-Tolerant Computing, pp. 422–443. IEEE Press, Toulouse (1993)
Hou, R.Y., Menon, J., Patt, Y.N.: Balancing I/O Response Time and Disk Rebuild Time in a RAID5 Disk Array. In: 26th Hawaii Int’l Conf. on Systems Sciences, pp. 70–79. IEEE Press, Hawaii (1993)
Lee, J.Y.B., Lui, J.C.S.: Automatic Recovery from Disk Failure in Continuous-Media Servers. J. IEEE Transaction on Parallel and Distributed Systems 13(5), 499–515 (2002)
Wu, S.Z., Feng, D., Jiang, H., Mao, B., Zeng, L.F., Chen, J.: JOR: A Journal-Guided Reconstruction Optimization for RAID Structured Storage Systems. In: 15th Int’l Conf. on Parallel and Distributed Systems, pp. 609–616. IEEE Press, Shenzhen (2009)
Wu, S.Z., Jiang, H., Mao, B.: IDO: Intelligent Data Outsourcing with Improved RAID Reconstruction Performance in Large-Scale Data Centers. In: 26th Int’l Conf. on Large Installation System Administration, pp. 17–32. USENIX Association, San Diego (2012)
Wan, S., Cao, Q., Huang, J.Z., Li, S.Y., Li, X., Zhan, S.H., Yu, L., Xie, C.S., He, X.B.: Victim Disk First: An Asymmetric Cache to Boost the Performance of Disk Arrays Under Faulty Conditions. In: USENIX Annual Technical Conference. USENIX Association, Berkeley (2011)
Gomez, M.E., Sontonja, V.: Characterizing Temporal Locality in I/O Workload. In: Int’l Symp. on Performance Evaluation of Computer and Telecommunication Systems, San Diego (July 2002)
SPC, Storage Performance Council I/O Traces, http://traces.cs.umass.edu/index.php/Storage/Storage
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2013 Springer International Publishing Switzerland
About this paper
Cite this paper
Liu, F., Pan, W., Xie, T., Gao, Y., Ouyang, Y. (2013). PDB: A Reliability-Driven Data Reconstruction Strategy Based on Popular Data Backup for RAID4 SSD Arrays. In: Kołodziej, J., Di Martino, B., Talia, D., Xiong, K. (eds) Algorithms and Architectures for Parallel Processing. ICA3PP 2013. Lecture Notes in Computer Science, vol 8285. Springer, Cham. https://doi.org/10.1007/978-3-319-03859-9_7
Download citation
DOI: https://doi.org/10.1007/978-3-319-03859-9_7
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-03858-2
Online ISBN: 978-3-319-03859-9
eBook Packages: Computer ScienceComputer Science (R0)