Update serializability in locking

  • R. C. Hansdah
  • L. M. Patnaik
Contributed Papers
Part of the Lecture Notes in Computer Science book series (LNCS, volume 243)


There are many concurrency control algorithms that use weak consistency of read-only transactions to enhance concurrency in database systems[3,8,18]. We identify these schedules (which become serializable when the weakly consistent read-only transactions are deleted from them) as update serializable schedules. In this paper, we study how update serializable schedules can be produced using locking and the advantages of allowing update serializable schedules in locking protocols. We show that if a locking protocol P which uses exclusive mode locks only is made heterogeneous, then the new locking protocol, called the heterogeneous locking protocol P, ensures update serializability. In addition, we also show that if the original locking protocol is deadlock-free, then a deadlock cycle in the heterogeneous locking protocol involves at least two read-only transactions. We also discuss the advantages that one gets when the guard locking protocol is made heterogeneous. The heterogeneous two phase locking protocol operates on general databases and ensures serializability. We present a simple non-two phase locking protocol on general databases that produces update serializable schedules which are not serializable. This protocol allows the read-only transactions to be non-two phase and restricts the update transactions to access exactly one data item.


Data Item Directed Acyclic Graph Hold Time General Database Root Vertex 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Buckley, G.N., Silberschatz, A. On the heterogeneous guard locking protocol. The Computer Journal,27,1(1984), pp.86–87.Google Scholar
  2. 2.
    Buckley, G.N., and Silberschatz, A. Beyond two phase locking. J. ACM, 32,2(Apr. 1985),314–326.Google Scholar
  3. 3.
    Chan, A., and Gray, R. Implementing distributed read-only transactions. IEEE Trans. Soft. Eng. SE-11, 2(Feb. 1985),205–212.Google Scholar
  4. 4.
    Casanova, M.A. The concurrency control problem for database systems. Lecture Notes in Computer Science, vol. 116,Berlin-Heidelberg-New York, Springer 1981.Google Scholar
  5. 5.
    Dasgupta,P., and Kedem,Z.M. A non-2-phase locking protocol for general databases. Proc. of the 8th Int. Conference on Very Large Databases, Oct. 1983, 92–96.Google Scholar
  6. 6.
    Eswaran, K.P., Gray, J.N., Lorie, R.A.,and Traiger, I.L. The notions of consistency and predicate locks in a database system. Comm. ACM, 19,11(Nov. 1976),624–633.Google Scholar
  7. 7.
    Garcia-Molina, H. Using semantic knowledge for transaction processing in distributed databases. ACM Trans. Database Syst., 8, 2(1983),186–213.Google Scholar
  8. 8.
    Garcia-Molina, H. Read-only transactions in distributed databases. ACM Trans. Database Syst., 7, 2(Jun. 1982),209–234.Google Scholar
  9. 9.
    Kedem, Z.M., and Silberschatz, A. Locking protocols from exclusive to shared locks. J. ACM, 30, 4(Oct.1983),787–804.Google Scholar
  10. 10.
    Korth,H. Locking protocols: general lock classes and deadlock freedom. Ph.D. Thesis, Princeton University, Jun. 1981.Google Scholar
  11. 11.
    Kung, H.T., and Papadimitriou, C.H. An optimality theory of concurrency control in databases. Acta Informatica, 19, 1(1983), 1–12.Google Scholar
  12. 12.
    Lamport,L. Towards a theory of correctness for multi-user database systems. Technical Report CA 7610-0712, Masachussetts Computer Associates Inc., Oct. 1976.Google Scholar
  13. 13.
    Mohan,C. Strategies for enhancing concurrency and managing deadlocks in database locking protocols. Ph.D. Thesis, Dept. of Computer Science, University of Texas at Austin, 1981.Google Scholar
  14. 14.
    Mohan, C., Fussel, D., and Silberschatz, A. Compatibility and commutativity of lock modes. Information and control,61,1(April 1984), 38–64.Google Scholar
  15. 15.
    Papadimitriou, C.H. The serializability of concurrent database updates. J. ACM, 29,4(Oct. 1979), 631–653.Google Scholar
  16. 16.
    Silberschatz, A., and Kedem, Z.M. Consistency in hierarchical database systems. J. ACM, 27,1(Jan. 1980), 72–80.Google Scholar
  17. 17.
    Silberschatz, A., and Kedem, Z.M. A family of locking protocols for database systems that are modeled by directed graphs. IEEE Trans. Soft. Engg., SE-8,6(Nov. 1982), 558–602.Google Scholar
  18. 18.
    Stearns,R.E., Lewis,P.M. II, and Rosenkrantz,D.Z. Concurrency control for database systems. Proc. 17th Symp. on Foundations of Computer Science, Oct.1976, 19–32.Google Scholar
  19. 19.
    Ullman, J.D. Principles of database systems. Computer Science Press Inc., Potomac, Md., 1982.Google Scholar
  20. 20.
    Yannakakis, M. Serializability by locking. J. ACM, 31, 2(Apr. 1984), 227–244.Google Scholar
  21. 21.
    Yannakakis, M. A theory of safe locking policies in database systems. J. ACM, 29,3(July 1982), 718–740.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1986

Authors and Affiliations

  • R. C. Hansdah
    • 1
  • L. M. Patnaik
    • 1
  1. 1.Department of Computer Science and AutomationIndian Institute of ScienceBangaloreIndia

Personalised recommendations