Abstract
Disk mirroring or RAID level 1 stores the same data twice, on two independent disks, to ensure that all single disk failures can be tolerated. This high storage overhead is acceptable in view of the drop in storage cost per gigabyte and rapidly increasing disk capacities. Disk access time, on the other hand, is improving at a very slow pace, so that another important advantage of disk mirroring is the doubling of the disk access bandwidth in processing read requests. Efficient routing of read requests to disks and local disk scheduling can be used to improve performance even further. We are primarily concerned with two RAID1 configurations: (i) source-initiated routing with the independent queues—IQ method; (ii) destination-initiated routing with the shared queue—SQ method. Static, dynamic, and affinity-based (AB) routing methods are used to distribute requests with the IQ method. We compare the performance of various IQ and SQ based routing policies using a random number-driven simulation. While there is some improvement in performance with the more sophisticated routing policies, performance is dominated by the local disk scheduling policy. The SQ method allows resource sharing, so that it tends to outperform the IQ based routing, but it requires the scheduler to keep track of the state of the disk drives. As a further means to improve performance, we consider the effect of prioritizing reads with respect to writes, transposed data allocation, and replicating data more than twice.
Similar content being viewed by others
References
D. Bitton and J. Gray, Disk shadowing, in: Proc. 24th Int’l Very Large Data Base Conf.—VLDB, Los Angeles, CA (August 1988) pp. 331–338.
P. M. Chen, E. K. Lee, G. A. Gibson, R. H. Katz, and D. A. Patterson, RAID: High-performance, reliable secondary storage, ACM Computing Surveys 26(2) (June 1994) 145–185.
S.Z. Chen and D.F. Towsley, A performance evaluation of RAID architectures, IEEE Trans. Computers 45(10) (October 1996) 1116– 1130.
E.G. Coffman Jr. and M. Hofri, Queueing models of secondary storage devices, in: Stochastic Analysis of Computer and Communication Systems, H. Takagi (Ed.) (Elsevier, 1990) pp. 549– 588.
J. Gray and P. Shenoy, Rules of thumb in data engineering, in: Proc. 16th Ann’l Int’l Conf. on Data Engineering—ICDE, San Diego, CA (March 2000) pp. 3–12.
M. Hofri, Should the two-headed disk be greedy? –Yes it should, Information Processing Letters—IPL 16(2) (1983) 83–85.
H.-I. Hsiao and D.J. DeWitt, A performance study of three high availability data replication strategies, J. Distr. Parallel Databases 1(1) (1993) 53–80.
R. P. King, Disk arm movement in anticipation of future requests, ACM Trans. Computer Systems 8(3) (1990) 214–229.
L. Kleinrock, Queueing Systems, Vol. 1/2: Theory/Computer Applications (Wiley, 1975/6).
R. W.-M. Lo and N. S. Matloff, A probabilistic limit on the virtual size of replicated disk systems, IEEE Trans. Knowledge and Data Eng. 4(1) (1992) 99–102.
C.R. Lumb, J. Schindler, and G.R. Ganger, Freeblock scheduling outside of disk firmware, in: Proc. 1st Conf. File and Storage Technologies -FAST’02, Monterey, CA (January 2002) pp. 275– 288.
J. Menon and J. Cortney, The architecture of a fault-tolerant cached RAID controller, in: Proc. 20th Annual Int’l Symp. on Computer Architecture—ISCA, Santa Margherita Ligure, Italy (June 1993) pp. 76–86.
S.-W. Ng, Reliability, availability and performance analysis of duplex disk systems, Reliability and Quality Control, M. H. Hamza (Ed.) ACTA Press (1987) pp. 5–9.
S.W. Ng, Improving disk performance via latency reduction, IEEE Trans. Computers 40(1) (1991) 22–30.
S.-W. Ng, Advances in disk technology: Performance issues, IEEE Computer 40(1) (May 1998) 75–81.
C. Orji and J. A. Solworth, Doubly distorted mirrors, in: Proc. ACM Int’l SIGMOD Conf Conf. on Management of Data, Washington, D. C. (May 1993) pp. 307–316.
C. Orji, M.A. Weiss, and J. A. Solworth, Improved traditional mirrors, in: Proc. 4th Int’l Conf. on Foundations of Data Organization and Algorithms (FODO 1993), Chicago, IL (October 1993), Lecture Notes in Computer Science, LNCS 730, David B. Lomet (Ed.), Springer-Verlag (1993) pp. 329–344.
C. Polyzois, A. Bhide, and D. M. Dias, Disk mirroring with alternating deferred updates, in: Proc. 19th Int’l Conf. on Very Large Data Bases—VLDB, Dublin, Ireland (August 1993) pp. 604– 617.
K.K. Ramakrishnan, P. Biswas, and R. Karedla, Analysis of file I/O traces in commercial computing environments, in: Proc. ACM SIGMETRICS Conf. on Measurement and Modeling of Computer Systems, Newport, RI (June 1992) pp. 78–90.
J.A. Solworth and C.U. Orji, Write-only disk caches, in: Proc. ACM Int’l SIGMOD Conf. on Management of Data, Atlantic City, NJ (May 1990) pp. 123–132.
J.A. Solworth and C.U. Orji, Distorted mirrors, in: Proc. First Int’l Conf. on Parallel and Distributed Information Systems—PDIS, Miami Beach, FL (December 1991) pp. 10–17.
Tandem Database Group. NonStop SQL: A distributed, high-performance, high-reliability implementation of SQL, High Performance Transaction Systems, D. Gawlick et al. (Ed.) (Springer-Verlag, 1987) pp. 60–103.
Teradata Corporation. DBC/1012 Database Computer System Manual, Release 2 (November 1985).
A. Thomasian and C. Liu, Some new disk scheduling policies and their performance, in: Proc. ACM SIGMETRICS Int’l Conf. on Measurement and Modeling of Computer Systems, Marina del Rey, CA (June 2002) pp. 266–267.
A. Thomasian, J. Spirollari, C. Liu, C. Han, and G. Fu, Mirrored disk scheduling, in: Proc. Int’l Symp. in Performance Evaluation of Computer and Telecommunication Systems, Montreal, Quebec, Canada (July 2003) pp. 580–584.
A. Thomasian and C. Han, Affinity based routing in zoned mirrored disks, The Computer Journal 48(3) (March 2005) pp. 292–299.
A. Thomasian and C. Liu, Comment on Issues and challenges in the performance analysis of real disk arrays, IEEE Trans. Parallel and Distributed Systems—TPDS 16(11) (November 2005) pp. 1102–1103.
A. Thomasian and C. Liu, Performance comparison of mirrored disk scheduling methods with a shared non-volatile cache, J. Distributed and Parallel Databases—JDPD 18(3) (December 2005) pp. 253–281.
A. Thomasian and G. Fu, Anticipatory disk arm placement to reduce seek time, Computer Systems: Science and Engineering—CSSE 21(3) (May 2006).
D.F. Towsley, S. Chen, and S.-P. Yu, Performance analysis of fault-tolerant mirrored disk systems, in: Performance 1990 (North-Holland, 1990) pp. 239–253.
A. Vakali and Y. Manolopoulos, An exact analysis on expected seeks in shadowed disks, Information Processing Letters (IPL) 61(6) (1997) 323–329.
A. Vakali and Y. Manolopoulos, Data placement schemes in replicated mirrored disk systems, The Journal of Systems and Software 55 (2000) 115–128.
Y.-T. Wang and R. J. T. Morris, Load sharing in distributed systems, IEEE Trans. Computers 34(3) (1985) 204–217.
B. L. Worthington, G. R. Ganger, and Y. L. Patt, Scheduling for modern disk drives and non-random workloads, in: Proc. ACM SIGMETRICS Conf. on Measurement and Modeling of Computer Systems, Nashville, TN (May 1994) pp. 241–251.
X. Yu, B. Gum, Y. Chen, R. Y. Wang, K. Li, A. Krishnamurthy, and T. E. Anderson, Trading capacity for performance on a disk array, in: Proc. 4th Symp. on Operating System Design and Implementation, San Diego, CA (October 2000) pp. 243–258.
Author information
Authors and Affiliations
Corresponding author
Additional information
We acknowledge the support of NSF through Grant 0105485 in Computer Systems Architecture.
Alexander Thomasian (ACM) is a Professor of Computer Science at New Jersey Institute of Technology. He was a faculty member at Case Western University and University of Southern California and an adjunct professor at Columbia University, while a Research Staff Member at the IBM T. J. Watson Research Center (1985–1998). He received his PhD in Computer Science from UCLA. Dr. Thomasian’s research has more recently been focused on indexing high-dimensional datasets and especially the performance of storage systems. He is the author of Database Concurrency Control: Methods, Performance, and Analysis, Kluwer 1996. Dr. Thomasian is a Fellow of IEEE.
Rights and permissions
About this article
Cite this article
Thomasian, A. Mirrored disk rouing and scheduling. Cluster Comput 9, 475–484 (2006). https://doi.org/10.1007/s10586-006-0014-3
Received:
Revised:
Accepted:
Issue Date:
DOI: https://doi.org/10.1007/s10586-006-0014-3