A Method for RAID Availability Analysis Based on Bernoulli Trials

Abstract:

With the processing speeds of processors increasing rapidly, the inefficiency of disk I/O performance has been becoming the bottleneck of computer systems and network transmission. So, Redundant Arrays of Independent (or Inexpensive) Disks (RAID) with the purpose of promoting disks parallel accessing , disks space utilization and user data availability has been adopted by DAS and SAN, and has been applied widely. However, the more quantity and capacity that disks have, the more mistakes that would take place, so RAID availability analysis has become an important and urgent issue. The paper proposed a set of availability computational modeling for RAID storage system via Bernoulli trials and binomial distribution. On the basis of it, it can be analyzed that combination orders and methods have an effect on RAID availability in the state of mixed combination; it also can be analyzed that disk quantity is associated with RAID availability. Both of them provide powerful decision support and theoretical foundation for user selecting RAID construction solution, and help users save cost of storage.

• Periodical: Applied Mechanics and Materials (Volumes 701-702)
• Pages: 129-135
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.701-702.129
• Online since: December 2014
• Authors: Li Qiang Zhu*
• Keywords: Bernoulli Trial, Degraded Mode, Mirroring Redundancy, Parity-Check Redundancy, RAID Availability Modeling

References

[1] Wang shuguang, Lv Guoren. RAID and its Applications[R]. Jilin: Jilin communications Polytechnic, (2009).

[2] Luo Haizhen. Applications of RAID techenology in windows NT[J]. Silicon Valley, 2009, (20): 146.

[3] HAN Chao, NA Wenwu, LIU Zhenjun, et al. Research on BW-RAID reconstruction technology and optimization[J]. Computer Engineering and Applications, 2012, 48(1): 60-61.

[4] Du Yunyun. A cloud storage reliability mechanism for erasure coding technique[J]. Computer Applications and Software, 2014, 31(2): 312-315.

[5] Tom Clark. Designing storage area networks[M]. Beijing: Electronic Industry Press, (2005).

[6] poloyzhang. RAID 0[EB/OL]. [2009-3-2]. http: /www. uplook. cn/blog/9/99883.

[7] Fang Liang, Lu Xicheng. A Cost Effective Fault-Tolerant Scheme for RAIDs[J]. Journal of Computer Science and Technology, 2003, 18(2): 230-234.

DOI: 10.1007/bf02948889

[8] Jackie. chen. A brief description of fault-tolerant principle of RAID3, 5, 6. [EB/OL]. [2009-3-17]. http: /jackiechen. blog. 51cto. com/196075/139615.

[9] Shi Qingsheng, Chen Xiaolong, Deng Xiaowei, et al. Probability and Mathematical Statistics[M]. Beijing: Chemical Industry Press, 2008: 30-33.

[10] Wang Shujiang, Liu Xiaohui. SAN, data backup and its recovery. [M]. Beijing: Electronic Industry Press, 2010: 20.

[11] Michael Gilroy, James Irvine. RAID6 Hardware Acceleration[J]. ACM Transactions on Embedded Computing Systems, 2011, 10(4).

[12] Liu Peng, Li Sanli, Francis C.M. Lau, Shi Yao, Huang Feng. RAID-M: A high performance RAID Matrix mass storage[J]. Science in China. Series f: Information Sciences, 2005, 48(4): 413 -420.

DOI: 10.1360/04yf0060

[13] Wu Suzhen, Chen Xiaoxi, Mao Bo. GC-RAIS: Garbage Collection Aware and Redundant Array of Independent SSDs[J]. Journal of Computer Research and Development, 2013, 50(1): 60 -63.