This Article 
   
 Share 
   
 Bibliographic References 
   
 Add to: 
 
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb
 
 Search 
   
Voting as the Optimal Static Pessimistic Scheme for Managing Replicated Data
January 1994 (vol. 5 no. 1)
pp. 64-73

This paper investigates the problem of finding an optimal static pessimistic replica control scheme. It has been widely accepted that coteries (proposed by Garcia-Molina and Barbara) provide the most general framework for such schemes. We demonstrate that voting schemes, a very small subset of static pessimistic schemes, are optimal for fully connected networks with negligible link failure rates, as well as for Ethernet systems. Wealso show that voting is not optimal for somewhat more general systems. We propose amodification of the algorithm of Z. Tong and R.Y. Kain (1988) for computing optimal voting in operation independent case, so that it runs in linear (rather than exponential) time. Finally, we propose the first efficient algorithm for computing the optimal vote assignment and appropriate thresholds for fully connected networks when relative frequencies of read and write operations are known. We also extend this result to Ethernet systems.

[1] M. Ahamad and M. H. Ammar, "Performance characterization of quorum-consensus algorithms for replicated data," inProc. IEEE Conf. Reliability in Distributed Software and Database Systems, pp. 161-167, 1987.
[2] P.A. Bernstein, V. Hadzilacos, and N. Goodman,Concurrency Control and Recovery in Database Systems, Addison-Wesley, Reading, Mass., 1987.
[3] P. A. Bernstein and N. Goodman, "An algorithm for concurrency control and recovery in replicated distributed databases,"ACM Trans. Database Syst., vol. 9, no. 4, pp. 596-615, Dec. 1984.
[4] S. Y. Cheunget al., "Optimizing vote and quorum assignments for reading and writing replicated data," inProc. Fifth IEEE Int. Conf. Data Eng., Los Angeles, CA, Jan. 1989.
[5] S. Y. Cheunget al., "Multi-dimensional voting: A general method for implementing synchronization in distributed systems," inProc. Tenth Int. Conf. Distributed Comput. Syst., Paris, June 1990.
[6] S.B. Davidson, H. Garcia-Molina, and D. Skeen, "Consistency in a Partitioned Network,"ACM Computing Surveys, Vol. 17, No. 3, Sept. 1985, pp. 341-370.
[7] H. Garcia-Molina and D. Barbara, "Optimizing the reliability provided by voting mechanisms," inProc. 4th Int. Conf. Distributed Computing Systems, pp. 340-346, 1984.
[8] H. Garcia-Molina and D. Barbara, "How to assign votes in a distributed systems,"J. ACM, vol. 32, no. 4, pp. 841-860, Oct. 1985.
[9] D. Barbara, H. Garcia-Molina, and A. Spauster, "Protocols for dynamic vote reassignment," inProc. ACM Conf. Principles Distributed Comput., 1986, pp. 195-205.
[10] D. Gifford, "Weighted voting for replicated data," inProc. 7th ACM Symp. Oper. Syst. Principles, Dec. 1979, pp. 150-162.
[11] M. P. Herlihy, "A quorum-consensus replication method for abstract data types,"ACM Trans. Comput. Syst., vol. 4, no. 1, Feb. 1986.
[12] S. Jajodia and D. Mutchler, "Enhancements to the voting algorithm," inProc. 13th Int. Conf. Very Large Data Bases, Sept. 1987, pp. 399-406.
[13] S. Jajodia and D. Mutchler, "Dynamic voting," inProc. 1987 ACM SIGMOD, San Francisco, CA, May 1987, pp. 227-238.
[14] S. Jajodia and D. Mutchler, "Integrating static and dynamic voting protocols to enhance file availability," inProc. 4th IEEE Int. Conf. Data Eng., 1988, pp. 144-153.
[15] S. Jajodia and D. Mutchler, "Dynamic voting algorithms for maintaining consistency of a replicated database,"ACM Trans. Database Syst., vol. 15, no. 2, June 1990.
[16] M. Obradovic and P. Berman, "Voting as the optimal static pessimistic scheme for managing replicated data," inProc. 9th Symp. Reliable Distributed Syst., 1990, pp. 126-135.
[17] M. Obradovic and P. Berman, "Weighted voting for operation dependent management of replicated data," inProc. 4th Int. Workshop on Distributed Algorithms, Lecture Notes in Computer Science. New York: Springer, 1990, pp. 263-276.
[18] M. Spasojevic, "Optimal replica control protocols for ring networks," inProc. 11th IEEE Symp. Reliable Distributed Syst., 1992, pp. 57-65.
[19] J. F. Paris and D. Long, "Efficient dynamic voting," inProc. Fourth IEEE Int. Conf. Data Eng., Los Angeles, CA, Jan. 1988.
[20] A. Rosenthal, "Computing the reliability of a complex network,"SIAM J. Appl. Math., vol. 32, pp. 384-393, 1977.
[21] J. Tang and N. Natarajan, "A static pessimistic scheme for handling replicated databases," inProc. 1989 ACM SIGMOD Int. Conf. Management Data, June 1989, pp. 389-398.
[22] J. Tang and N. Natarajan, "A static pessimistic scheme for managing replicated databases," Dept. of Computer Science, Penn State Univ., Tech. Rep., Feb. 1990.
[23] R. Thomas, "A majority consensus approach to concurrency control,"ACM Trans. Database Syst., vol. 4, pp. 180-209, June 1979.
[24] Z. Tong and R. Kain, "Vote assignments in weighted voting mechanisms," inProc. 7th Symp. Reliable Distributed Syst., 1988, pp. 138-143.

Index Terms:
Index Termsdistributed databases; software reliability; protocols; local area networks; optimal staticpessimistic scheme; replicated data; voting schemes; Ethernet systems; optimal voteassignment; fully connected networks
Citation:
M. Spasojevic, P. Berman, "Voting as the Optimal Static Pessimistic Scheme for Managing Replicated Data," IEEE Transactions on Parallel and Distributed Systems, vol. 5, no. 1, pp. 64-73, Jan. 1994, doi:10.1109/71.262589
Usage of this product signifies your acceptance of the Terms of Use.