MS-RAID: An Energy-Saving Data Layout for CDP
Continuous data protection (CDP) provides unlimited granular recovery point objective (RPO) and nearly instant recovery time objective (RTO). It requires a great fluctuations in the performance of storage system. The system requires higher storage bandwidth when it is active, and lower when inactive. Raid, that is used in storage system normally and provides fixed performance, may face a performance bottleneck or high power consumption. This paper proposes MS-RAID, that bases on S-Raid and can provide multi-level dynamic mapping storage scheme. MS-RAID has vary levels of grouping strategies. MS-RAID can meet the needs of real-time dynamic load by changing the number of parallel disks. When the throughput rises, MS-RAID turns the high-level disk group into running to avoid the bottleneck of system performance. When the throughput falls, MS-RAID turns the low-level disk group into running to save energy consumption. Experiments show that MS-RAID is a more energy-efficient data layout, and can save more energy consumption and improve then performance than S-RAID.
KeywordsRAID CDP Energy-saving Storage Data layout
The work was supported by the Natural Science Foundation of China (No. 61876019), the Natural Science Foundation of Hebei Province (Grant No. F2016202145), the Youth Foundation of Education Commission of Hebei Province (Grant No. QN2014192), and the Science and Technology Planning Project of Hebei Province of China (grant No. 15210325).
- 3.Dong, Y., Liu, J., Yang, J., et al.: HS-RAID 2: optimizing small write performance in HS-RAID. J. Electr. Comput. Eng. 2016, Article no. 7341735, 8 pages (2016)Google Scholar
- 4.Patterson, D.: A case for redundant arrays of inexpensive disks. In: Proceedings of ACM SIGMOD Conference (1988)Google Scholar
- 5.Gurumurthi, S., Sivasubramaniam, A., Kandemir, M., et al.: DRPM: dynamic speed control for power management in server class disks. In: Proceedings of the 30th Annual International Symposium on Computer Architecture, San Diego, pp. 169–179. IEEE (2003)Google Scholar
- 6.Sun, Z., Zhang, Q., Li, Y., Tan, Y.A., et al.: DPPDL: a dynamic partial-parallel data layout for green video surveillance storage. IEEE Trans. Circuits Syst. Video Technol. PP(99), 1 (2016)Google Scholar
- 7.Xu, L., Cipar, J., Krevat, E., et al.: SpringFS: bridging agility and performance in elastic distributed storage. In: Proceedings of Usenix Conference on File and Storage Technologies, pp. 243–255. USENIX Association (2014)Google Scholar
- 8.Basmadjian, M., Hermann, M.D., Lent, R., Giovanni, G.: Cloud computing and its interest in saving energy: the use case of aprivate cloud. J. Cloud Comput. Adv. Syst. Appl. 1(5), 1–11 (2012)Google Scholar
- 9.Eric, S., Michael, L., Jon, S., et al.: Computational solutions to large-scale data management and analysis. Nat. Rev. Genet. 11, 647–657 (2010)Google Scholar
- 10.Li, X., Tan, Y., Sun, Z.: Semi-RAID: a reliable energy-aware RAID data layout for sequential data access. In: Proceedings of IEEE, Symposium on MASS Storage Systems and Technologies, pp. 1–11. IEEE Computer Society (2011)Google Scholar
- 11.Liu, J., Zhang, J., Li, Y., et al.: Hybrid S-RAID: an energy-efficient data layout for sequential data storage. J. Comput. Res. Dev. 50(1), 37–48 (2013). (in Chinese)Google Scholar
- 12.Liu, J., Tan, Y., Xue, J., et al.: Writing optimization strategy in S-RAID based on sequential data characteristics. Chin. J. Comput. 37(3), 721–734 (2014). (in Chinese)Google Scholar