Abstract
In the era of big data, the traditional RAID storage system has been incapable of meeting the requirements of performance and reliability for the large amount of data storage and computing. In view of the situation, Solid State Disks (SSDs), which can provide better performance than Hard Disk Drives (HDDs), are widely used to construct storage arrays in enterprise environments. Today many studies on Redundant Array of Independent SSDs (RAIS) storage systems concentrate more on improving write performance, and show less attention on the reconstruction performance of RAIS storage systems. In this paper, we proposed RAIS5AS, a novel RAIS5 storage architecture with adaptive stripe for improving the performance and reliability of RAIS5. RAIS5AS distinguishes between logical stripe and physical stripe. Logical stripe is a traditional RAID stripe. Physical stripe consists of the blocks (in a logical stripe) which have been written data. When handling write requests, RAIS5AS uses physical stripe as the basic processing unit to choose which blocks are read to compute the new parity block. When recovering data, RAIS5AS skips the unused failed blocks. In addition, RAIS5AS simplifies the synchronization process of RAIS5 storage system. We have implemented the proposed scheme and carried out a series of experiments. RAIS5AS on average improves write performance and reconstruction performance of the basic RAIS5 by up to 7.92% and 95.65% respectively, and those of JOR by 9.16% and 14.57% respectively.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Mei, L., Feng, D., Zeng, L., Chen, J., Liu, J.: A write-through cache method to improve small write performance of SSD-based RAID. In: Proceedings of the 12th International Conference on Networking, Architecture, and Storage, Shenzhen, China, pp. 1–6, August 2017
Kim, T., Lee, S., Park, J., Kim, J.: Efficient lifetime management of SSD-based RAIDs using dedup-assisted partial stripe writes. In: Proceedings of the 5th Non Volatile Memory Systems and Applications Symposium, Daegu, South Korea, pp. 1–6, August 2016
Wan, J., Wu, W., Zhan, L., Yang, Q., Qu, X., Xie, C.: DEFT-cache: a cost effective and highly reliable SSD cache for RAID storage. In: Proceedings of the 2017 IEEE International Parallel and Distributed Processing Symposium, Orlando, FL, pp. 102–111, May 2017
Lee, E., Oh, Y., Lee, D.: SSD caching to overcome small write problem of disk-based RAID in enterprise environments. In: Proceedings of the 30th Annual ACM Symposium on Applied Computing Software Verification and Testing Track, Salamanca, Spain, pp. 2047–2053, April 2015
Pan, Y., Li, Y., Xu, Y., Shen, B.: DCS: diagonal coding scheme for enhancing the endurance of SSD-based RAID arrays. IEEE Trans. Comput.-Aided Des. Integr. Circ. Syst. 35(8), 1372–1385 (2016)
Wu, S., Mao, B., Chen, X., Jiang, H.: LDM: log disk mirroring with improved performance and reliability for SSD-based disk arrays. ACM Trans. Storage (TOS) 12(4), 22 (2016)
Tian, L., et al.: PRO: a popularity-based multi-threaded reconstruction optimization for RAID-structured storage systems. In: Proceedings of the 2007 USENIX Conference on File and Storage Technologies, San Jose, CA, pp. 277–290, February 2007
Sivathanu, M., Prabhakaran, V., Arpaci-Dusseau, A., Arpaci-Dusseau, R.: Improving storage system availability with D-GRAID. ACM Trans. Storage (TOS) 1(2), 133–170 (2005)
Wu, S., Feng, D., Jiang, H., Mao, B., Zeng, L., Chen, J.: JOR: a journal-guided reconstruction optimization for RAID-structured storage systems. In: Proceedings of the 15th IEEE International Conference on Parallel and Distributed Systems, Shenzhen, China, , pp. 609–616, December 2009
Jin, C., Feng, D., Jiang, H., Tian, L.: RAID6L: a log-assisted RAID6 storage architecture with improved write performance. In: Proceedings of the IEEE 27th Symposium on Mass Storage Systems and Technologies, Denver, Colorado, pp. 1-6, May 2011
UMass Trace Repository: Two I/O traces from OLTP applications running at two large financial institutions (2007). http://traces.cs.umass.edu/index.php/Storage/Storage
Microsoft Enterprise Traces (2009). http://iotta.snia.org/traces/list/BlockIO
Agrawal, N., Prabhakaran, V., Wobber, T., Davis, J., Manasse, M., Panigrahy, R.: Design tradeoffs for SSD performance. In: Proceedings of the 2008 USENIX Annual Technical Conference, Boston, MA, pp. 57–70, June 2008
Wu, S., Jiang, H., Feng, D., Tian, L., Mao, B.: WorkOut: I/O workload outsourcing for boosting RAID reconstruction performance. In: Proceedings of the USENIX Conference on File and Storage Technologies, San Francisco, CA, pp. 239–252, February 2009
Holland, M., Gibson, G., Siewiorek, D.: Architectures and algorithms for on-line failure recovery in redundant disk arrays. J. Distrib. Parallel Databases 2(3), 295–335 (1994)
Holland, M., Gibson, G., Siewiorek, D.: Fast, on-line failure recovery in redundant disk arrays. In: Proceedings of the 23 Annual International Symposium on Fault-Tolerant Computing, Toulouse, France, pp. 422–431, June 1993
Lee, J., Lui, J.: Automatic recovery from disk failure in continuous-media servers. IEEE Trans. Parallel Distrib. Syst. 13(5), 499–515 (2002)
Mao, B., et al.: a hybrid parity-based disk array for enhanced perfromance and reliability. ACM Trans. Storage 8(1), 1C20 (2012)
Zeng, L., Feng, D., Chen, J., Wei, Q.: HRAID6ML: a hybrid RAID6 storage architecture with mirrored logging. In: Proceedings of the IEEE 28th Symposium on Mass Storage Systems and Technologies, Pacific Grove, CA, pp. 1–6, April 2012
Acknowledgements
This work was supported by the National High Technology Research and Development Program (863 Program) No. 2015AA016701, No. 2015AA015301; NSFC No. 61772222, No. U1705261, No. 61472153, No. 61402189, No. 61303046; State Key Laboratory of Computer Architecture, No. CARCH201505; Wuhan Applied Basic Research Project (No. 2015010101010004).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer Nature Switzerland AG
About this paper
Cite this paper
Mei, L., Feng, D., Zeng, L., Chen, J., Liu, J. (2018). A High-Performance and High-Reliability RAIS5 Storage Architecture with Adaptive Stripe. In: Vaidya, J., Li, J. (eds) Algorithms and Architectures for Parallel Processing. ICA3PP 2018. Lecture Notes in Computer Science(), vol 11334. Springer, Cham. https://doi.org/10.1007/978-3-030-05051-1_39
Download citation
DOI: https://doi.org/10.1007/978-3-030-05051-1_39
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-05050-4
Online ISBN: 978-3-030-05051-1
eBook Packages: Computer ScienceComputer Science (R0)