This Article 
   
 Share 
   
 Bibliographic References 
   
 Add to: 
 
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb
 
 Search 
   
A Performance Evaluation of RAID Architectures
October 1996 (vol. 45 no. 10)
pp. 1116-1130

Abstract—In today's computer systems, the disk I/O subsystem is often identified as the major bottleneck to system performance. One proposed solution is the so-called redundant array of inexpensive disks (RAID). In this paper, we examine the performance of two of the most promising RAID architectures, the mirrored array and the rotated parity array. First, we propose several scheduling policies for the mirrored array and a new data layout, group-rotate declustering, and compare their performance with each other and in combination with other data layout schemes. We observe that a policy that routes reads to the disk with the smallest number of requests provides the best performance, especially when the load on the I/O system is high. Second, through a combination of simulation and analysis, we compare the performance of this mirrored array architecture to the rotated parity array architecture. This latter study shows that, 1) given the same storage capacity (approximately double the number of disks), the mirrored array considerably outperforms the rotated parity array, and 2) given the same number of disks, the mirrored array still outperforms the rotated parity array in most cases, even for applications where I/O requests are for large amounts of data. The only exception occurs when the I/O size is very large; most of the requests are writes, and most of these writes perform full stripe write operations.

[1] C.G. Bell, "The Mini and Micro Industries," Computer, vol. 17, no. 10, pp. 14-30, Oct. 1984.
[2] D. Bitton and J. Gray, “Disk Shadowing,” Very Large Data Bases, pp. 331–338, 1988.
[3] D. Bitton, "Arm Scheduling in Shadowed Disks," Proc. IEEE Spring ComCon, pp. 132-136, Feb. 1989.
[4] D.L. Bultman, "High Performance SCSI Using Parallel Drive Technology," Proc. BUSCON Conf., pp. 40-44,Anaheim, Calif., Feb. 1988.
[5] P. Chen, G. Gibson, R.H. Katz, and D.A. Patterson, "An Evaluation of Redundant Arrays of Disks Using an Amdahl 5890," Performance Evaluation Rev., vol. 18, no. 1, pp. 74-85, May 1990.
[6] S.-Z. Chen, "Design, Modeling, and Evaluation of High Performance I/O Subsystems," PhD thesis, Univ. of Massachusetts at Amherst, Sept. 1992.
[7] S.-Z. Chen and D. Towsley, "Raid 5 vs. Parity Striping: Their Design and Evaluation," J. Parallel and Distributed Computing, vol. 17, pp. 58-74, 1992.
[8] G. Copeland and T. Keller, “A Comparison of High-Availability Media RecoveryTechniques,” Proc. ACM SIGMOD Conf., pp. 98–109, 1989.
[9] F. Douglis and J. Ousterhout, "Log-Structured File Systems," Proc. IEEE Spring ComCon, pp. 124-129, Feb. 1989.
[10] J. Gray, B. Horst, and M. Walker, "Parity Striping of Disk Arrays: Low Cost Reliable Storage with Acceptable Throughput," Proc. 16th Int'l VLDB Conf., p. 152, 1990.
[11] J.M. Harker et al., "A Quarter Century of Disk File Innovation," IBM J. Research and Development, vol. 25, pp. 677-689, Sept. 1981.
[12] H. Hsiao and D.J. DeWitt, “Chained Declustering: A New Availability Strategy for Multiprocessor Database Machines,” Proc. Data Eng., pp. 456–465, 1990.
[13] W. Joy Presentation at ISSCC '85 Panel Session, Feb. 1985.
[14] R. Katz, G. Gibson, and D. Patterson, “Disk System Architectures for High Performance Computing,” Proc. IEEE, vol. 77, no. 12, pp. 1,842–1,858, Dec. 1989.
[15] A.Y. Khinchine, Mathematical Methods in the Theory of Queueing.New York: Hafner Publishing, 1960.
[16] M.Y. Kim, "Synchronized Disk Interleaving," IEEE Trans. Computers, vol. 35, no. 11, pp. 978-988, Nov. 1986.
[17] M.Y. Kim and A.N. Tantawi, "Asynchronized Disk Interleaving: Approximating Access Delays," IEEE Trans. Computers, vol. 40, no. 7, pp. 801-810, July 1991.
[18] L. Kleinrock, Queueing Systems, vol. 1. New York: John Wiley, 1975.
[19] L. Kleinrock, Queueing Systems, vol. 2. New York: John Wiley, 1976.
[20] E.K. Lee and R.H. Katz, "Performance Consequences of Parity Placement in Disk Arrays," Proc. Fourth Int'l Conf. Architectural Support for Programming Languages and OS, pp. 190-199, Apr. 1991.
[21] E.K. Lee and R.H. Katz, "An Analytic Performance Model of Disk Arrays," Proc. 1993 SIGMETRICS Conf. Measurement and Modeling of Computer Systems, pp. 98-109,Santa Clara, Calif., 1993.
[22] M. Livny, S. Khoshafian, and H. Boral, “Multi-Disk Management Algorithms,” Proc. SIGMETRICS, pp. 69–77, May 1987.
[23] W.C. Lynch, "Do Disk Arms Move?" Performance Evaluation Rev., vol. 1, pp. 3-16, Dec. 1972.
[24] J. Menon and J. Kasson, "Methods for Improved Update Performance of Disk Arrays," Technical Report RJ 6928 (66034), IBM Almaden Research Center, Nov. 1990.
[25] J. Menon, D. Mattson, and S. Ng, "Performance of Disk Arrays in Transaction Processing Environments," Technical Report RJ 8230 (75424), IBM Almaden Research Center, July 1991.
[26] A. Merchant and P.S. Yu, "Modeling and Comparisons of Striping Strategies in Data Replication Architectures," technical report, IBM T.J. Watson Research Center, Sept. 1991.
[27] R.R. Muntz and J.C.S. Lui, "Performance Analysis of Disk Arrays Under Failure," Proc. 16th Int'l Conf. Very Large Data Bases, pp. 162-173,Brisbane, Australia, Aug. 1990.
[28] M.N. Nelson, B.B. Welch, and J.K. Ousterhout, "Caching in the Sprite File System," ACM Trans. Computer Systems, vol. 6, no. 1, pp. 134-154, Feb. 1988.
[29] S. Ng, "Some Design Issues of Disk Array," Proc. IEEE Spring ComCon, pp. 137-142, 1989.
[30] S. Ng, D. Lang, and R. Selinger, "Trade-Offs Between Devices and Paths in Achieving Disk Interleaving," Proc. 15th Int'l Symp. Computer Architecture, pp. 196-201,Hawaii, May 1988.
[31] M. Ogata and M. Flynn, "A Queueing Analysis for Disk Array Systems," Technical Report CSL-TR-90-443, Stanford Univ., Aug. 1990.
[32] J. Ousterhout and F. Douglis, Beating the I/O Bottleneck: A Case for Log-Structured File Systems Operating Systems Rev., vol. 23, no. 1, pp. 11-28, 1989.
[33] J.K. Ousterhout et al., "A Trace-Driven Analysis of the UNIX 4.2 BSD File System," Proc. 10th Symp. Operating Systems Principles, pp. 15-24, Dec. 1985.
[34] D.A. Patterson, P. Chen, G. Gibson, and R.H. Katz, “Introduction to Redundanct Array of Inexpensive Disks (RAID),” Proc. COMPCON '89, pp. 112-117, 1989.
[35] D.A. Patterson, G. Gibson, and R.H. Katz, “A Case for Redundant Arrays of Inexpensive Disks (RAID),” Proc. ACM SIGMOD Conf., pp. 109–116, 1988.
[36] A. Poston, "A High Performance File System for UNIX," Proc. USENIX, pp. 215-226, Sept. 1988.
[37] A. Reddy and P. Banerjee, “Evaluation of Multiple-Disk I/O Systems,” IEEE Trans. Computers, vol. 38, pp. 1,680–1,690, Dec. 1989.
[38] M. Rosenblum and J.K. Ousterhout, "The LFS Storage Manager," Proc. Summer '90 USENIX Technical Conf., pp. 315-324,Anaheim, Calif., June 1990.
[39] M. Rosenblum and J.K. Ousterhout, "The Design and Implemenation of a Log-Structured Fils System," draft, Mar. 1991.
[40] T. Ruwart and M. O'Keefe, "Performance Characteristics of a 100 Megabyte/Second Disk Array," presented in Storage&Interface '94, Jan. 1994.
[41] K. Salem and H. Garcia-Molina, Disk Striping Proc. Data Eng. '86, pp. 336-342, 1986.
[42] M. Schroeder, D. Gifford, and R. Needham, "A Caching File System for a Programmer's Workstation," Proc. 10th Symp. Operating System Principles, pp. 25-34, Dec. 1985.
[43] M. Schulze, G. Gibson, R. Katz, and D. Patterson, "How Reliable is a RAID?" Proc. IEEE Spring ComCon Conf., pp. 118-123,San Francisco, Feb. 1989.
[44] R.A. Scranton and D.A. Thompson, "The Access Time Myth," Technical Report RC 10197, IBM, Sept. 1983.
[45] M. Stonebraker and G.A. Schloss, “Distributed RAID—A New Multiple Copy Algorithm,” Proc. Sixth Int'l Conf. Data Eng., pp. 430-437, Feb. 1990.

Index Terms:
Disk arrays, I/O subsystems, performance analysis, queuing model, RAID, scheduling policies.
Citation:
Shenze Chen, Don Towsley, "A Performance Evaluation of RAID Architectures," IEEE Transactions on Computers, vol. 45, no. 10, pp. 1116-1130, Oct. 1996, doi:10.1109/12.543706
Usage of this product signifies your acceptance of the Terms of Use.