Peter A. Franaszek
John T. Robinson
Abstract
Future transaction processing systems may have substantially higher levels of concurrency due to reasons which include: (1) increasing disparity between processor speeds and data access latencies, (2) large numbers of processors, and (3) distributed databases. Another influence is the trend towards longer or more complex transactions. A possible consequence is substantially more data contention, which could limit total achievable throughput. In particular, it is known that the usual locking method of concurrency control is not well suited to environments where data contention is a significant factor.
Here we consider a number of concurrency control concepts and transaction scheduling techniques that are applicable to high contention environments, and that do not rely on database semantics to reduce contention. These include access invariance and its application to prefetching of data, approximations to essential blocking such as wait depth limited scheduling, and phase dependent control. The performance of various concurrency control methods based on these concepts are studied using detailed simulation models. The results indicate that the new techniques can offer substantial benefits for systems with high levels of data contention.
- 1 AGRAWAL, R., CAREY, M J., AND LIVNY, M. Concurrency control performance modeling: alternatives and implications ACM Trans. Database Syst. 12, 4 (Dec. 1987), 609-654. Google Scholar
Digital Library
- 2 BALTER, R., BERARD, P, AND DECITRE, P. Why control of concurrency level in distributed systems is more fundamental than deadlock management. In Proceedings of the ACM Symposium Principles of Distributed Computing (Aug. 1982) ACM, 1982, pp. 183-193. Google ScholarDigital Library
- 3 DuBouRDIEU, D.J. Implementation of distributed transactions. In Proceedings of the 6tb Berkeley Workshop on Distributed Data Management and Computer Networks, ACM/IEEE, 1982, pp. 81-94Google Scholar
- 4 FRANASZEK, P., AND ROBINSON, J.T. Limitations of concurrency in transaction processing. ACM Trans. Database Syst. 10, 1 (Mar. 1985), 1-28. Google ScholarDigital Library
- 5 GAWLIBK, D., AND KINKADE, D. Varieties of concurrency control in IMS/VS fast path IEEE Database Eng. 4 (1985), 63-70.Google Scholar
- 6 Hsu, M., AND ZHANG, B. Performance evaluation of cautious waiting Res. Rep., Aiken Computation Laboratory, Harvard Univ., Aug. 1987 (submitted for publication).Google Scholar
- 7 LAVENBERG, S. S. (ED), Computer Performance Modehng Handbook. Academic Press, San Diego, Calif., 1983. Google ScholarDigital Library
- 8 MALHOTRA, A., MARKOWITZ, H. M., AND PAZEL, D. P. Conmstent copy: a new database facility. Rep. RC9630, IBM Research Div., T. J. Watson Research Center, Yorktown Heights, N Y., 1982Google Scholar
- 9 O'NEIL, P E. The escrow transactional method. ACM Trans. Database Syst. 11, 4 (Dec. 1986), 405 430. Google ScholarDigital Library
- 10 REUTER, A Concurrency on high-traffic data elements. In Proceedzngs of the ACM Symposium on Principles of Database Systems, ACM, 1982, pp. 83-92. Google ScholarDigital Library
- 11 REUTER, A. The transaction pipeline processor. Presented at the International Workshop on High Performance Transactzon Systems (Pacific Grove, Calif, Sept. 1985).Google Scholar
- 12 ROmNSON, J. T Design of concurrency controls for transaction processing systems~ Ph.D. thesis, Tech. Rep. CMU-CS-82-114, Carnegie Mellon Univ , 1982.Google Scholar
- 13 RYu, I K., AND THOMASIAN, A. Performance analysis of centralized databases with optimistic concurrency control. Perform. Eval. 7, 3 (1987), 195 211. Google ScholarDigital Library
- 14 RYu, I. K., AND THOMASIAN, A. Performance analysis of dynamic locking with the no-waiting policy. {EEE Trans. Softw. Eng. 16, 7 (July 1990), 684 698. Google ScholarDigital Library
- 15 TAY, Y. C. Locking Performance in Centralized Databases. Academic Press, San Diego, Calif., 1987. Google ScholarDigital Library
- 16 THOMASIAN, A. Performance limits of two-phase locking. In Proceedings of the 7th International Conference of Data Engineering (Apr. 1991). IEEE, 1991 pp. 426-435. Google ScholarDigital Library
- 17 THOMASIAN, A., AND RYU, I.K. A decomposition solution to the queueing network model of the centralized DBMS with static locking. In Proceedings of the ACM SIGMETRICS Conference on Measurement and Modeling of Computer Systems (Aug. 1983). ACM, 1983, pp. 82-92. Google ScholarDigital Library
Index Terms
- Concurrency control for high contention environments
Recommendations
Analysis of Hybrid Concurrency Control Schemes for a High Data Contention Environment
Analytical models are developed to study hybrid CC (concurrency control) schemes which employ a different CC scheme to handle rerun transactions, since their characteristics are different from the first run of transactions. These include switching to ...
Real-time optimistic concurrency control protocol with dynamic adjustment of serialization order
RTAS '95: Proceedings of the Real-Time Technology and Applications SymposiumProposes a new real-time optimistic protocol. By using a dynamic adjustment of the serialization order by backward-adjusting the non-serious conflicting transactions before the committing transactions, many unnecessary restarts can be eliminated. In the ...
Semantics-based concurrency control: beyond commutativity
The concurrency of transactions executing on atomic data types can be enhanced through the use of semantic information about operations defined on these types. Hitherto, commutativity of operations has been exploited to provide enchanced concurrency ...
Reviews
Mukesh Singhal
The authors speculate that future data processing systems will have a substantially higher level of concurrency, and they argue that traditional locking methods are not suitable for locking in such environments because of high data contention. The authors present several concurrency control and transaction scheduling techniques for high contention environments. The primary techniques studied are running priority, wait-depth limited concurrency control, and access invariance. The authors have conducted a detailed performance study of the proposed methods using simulation. The study, however, is limited to centralized database systems with exclusive access. The main finding of the study is that wait-depth limited concurrency control is robust. The paper contains a detailed exposition of the material and is well organized. The paper is lengthy, however, and a background in concurrency control and CPU scheduling may be required. Overall, this paper is excellent for someone interested in techniques to achieve high throughput in an environment of severe data contention.
Access critical reviews of Computing literature here
Become a reviewer for Computing Reviews.
Comments