Abstract
Data availability is an important requirement of distributed databases. Replication is a technique that has been proposed to meet this need. In the absence of failures, traditional replica control algorithms provide complete availability in the sense that any transaction can be executed. The worst case of data availability occurs when the system is totally partitioned (each operational site is isolated from every other site). In this paper, we present techniques to achieve high availability under combinations of site failures and partitions. Users are required to specify the database access requirements in the totally-partitioned environment. This information is represented by means of a Read Access Graph (RAG). When failures occur, the set of items that may be accessed by a transaction depends on the connectivity of the network and the RAG. The techniques ensure that as failures occur the loss of availability is gradual and graceful. Data availability improves with the level of normalcy in the system. Unless there is a complete failure, at least some predefined set of transactions can be executed. It is shown that these algorithms preserve the integrity of the database by ensuring that all executions are one-copy serializable. The algorithms compare favorably with other replica management schemes in terms of availability.
Similar content being viewed by others
References
Bernstein PA, Goodman N: Concurrency control in distributed database systems. ACM Comput Surv 13(2): 185–221 (1981)
Bernstein PA, Goodman N: The failure and recovery problem for replicated dtabases. In: Proc 2nd Annu ACM Symp PODC, pp 114–122 (1983)
Bernstein PA, Goodman N, Hadzilacos V: Concurrency control and recovery in database systems. Addison-Wesley, Reading, Mass. 1987
Chan A, Gray R: Implementing distributed read-only transactions. IEEE Trans Software Eng, SE-11(2): 205–212 (1985)
Davidson SB, Garcia-Molina H, Skeen D: Consistency in partitioned networks. ACM Comput Surv 17(3): 346–370 (1985)
Eager DL, Sevcik KC: Achieving robustness in distributed database systems. ACM TODS, 8(3): 354–381 (1983)
El Abbadi A, Toueg S: Maintaining availability in partitioned replicated databases. ACM TODS 14(2): 264–290 (1989)
Eswaran KP, Gray JN, Lorie RA, Traiger IL: The notions of consistency and predicate locks in a database system. Commun ACM 19(11): 624–633 (1976)
Garcia-Molina H, Kogan B: Achieving high availability in distributed databases. IEEE Trans Software Eng 14(7): 886–896 (1988)
Garcia-Molina H, Kogan B: Update propagation in Bakunin data networks. In: Proc 6th ACM Symp PODC, pp 13–26, Vancouver, BC, 1987
Garcia-Molina H, Wiederhold G: Read-only transactions in a distributed database. ACM TODS 7(2): 209–234 (1982)
Gifford, DK: Weighted voting for replicated data. In: Proc 7th ACM SIGOPS Symp Operating Systems principles, pp 150–159, Pacific Grove, CA, 1979
Jajodia S, Mutchler D: Dynamic voting. In: Proc ACM SIGMOD Annu Conf, pp 227–238, San Francisco, 1987
Kung HT, Papadimitriou CH: An optimality theory of concurrency control for databases. Acta Inf 19: 1–13 (1983)
Long DDE, Paris JF: An realistic evaluation of optimistic dynamic voting. In: Proc 7th RDS Conf, pp 129–137, 1988
Minoura T, Wiederhold G: Resilient extended true-copy token scheme for a distributed database system. IEEE Trans Software Eng 8(3): 173–189 (1982)
Stonebraker M: Concurrency control and consistency of multiple copies of data in distributed ingres. IEEE Trans Software Eng 5(3): 188–194 (1979)
Thomas RE: A majority consensus approach to concurrency control for multiple copy databases. ACM TODS 4(2): 180–209 (1979)
Weihl WE: Distributed version managment for read-only actions. In: Proc 4th Annu ACM Symp PODC, pp 122–135, Minaki, Ontario, 1985
Author information
Authors and Affiliations
Additional information
K. Brahmadathan obtained a Bachelor's degree in Electronics and Communications Engineering from University of Kerala, Trivandrum, India; a Master's degree in Computer Science from Indian Institute of Technology, Madras, India; and the M.S. and Ph.D. degrees in Computer Science from University of Pittsburgh. Since 1989, he has been an Assistant Professor of Computer Science at the University of Wyoming. His research interests are in the areas of database systems and distributed systems.
K.V.S. Ramarao obtained his M.Sc. in Applied Mathematics from Andhra University, Waltair, India; M.Tech. in Computer Science from IIT Kanpur, India; and the Ph.D. in Computing Science from University of Alberta, Edmonton, Canada. He is currently a Senior Technologist for Southwestern Bell Technology Resources, Inc. Prior to that, he was an Assistant Professor at the University of Pittsburgh. His current research interests include distributed systems and distributed databases.
Rights and permissions
About this article
Cite this article
Brahmadathan, K., Ramarao, K.V.S. Achieving graceful performance in distributed error-prone databases. Distrib Comput 4, 163–174 (1991). https://doi.org/10.1007/BF01784718
Received:
Accepted:
Issue Date:
DOI: https://doi.org/10.1007/BF01784718