Abstract
Real-time database systems support applications which have severe performance constraints such as fast response time and continued operation in the face of catastrophic failures. Real-time database systems are still in the state of infancy, and issues and alternatives in their design are not very well explored. In this paper, we discuss issues in the design of real-time database systems and discuss different alternatives for resolving these issues. We discuss the aspects in which requirements and design issues of real-time database systems differ from those of conventional database systems. We discuss two approaches to design real-time database systems, viz., main memory resident databases and design by trading a feature (like serializability).
We also discuss requirements in the design of real-time distributed database systems, and specifically discuss issues in the design of concurrency control and crash recovery. It is felt that long communication delays may be a factor in limiting the performance of real-time distributed database systems. We present a concurrency control algorithm for real-time distributed database systems whose performance is not limited by communication delays.
- 1. ATTAR, R., BERNSTEIN, P. A., AND GOODMAN, N., "Site Initialization, Recovery, and Backup in a Distributed Database System," IEEE Trans. on Software Engineering , pp. 645-650 (November 1984).Google ScholarDigital Library
- 2. BITTON, DINA, "The Effect of Large Main Memory on Database Systems," Proc. of the ACM SIGMOD, pp. 337-339 (1986). Google ScholarDigital Library
- 3. BLAUSTEIN, B. T., GARCIA-MOLINA, H., RIES, D. R., CHILENSKAS, R. M., AND KAUFMAN, C. W., "Maintaining Replicated Databases Even in Presense of Network Partitions," EASCON, pp. 353-360 (1983).Google Scholar
- 4. CHAN, A., DAYAL, U., FOX, S., GOODMAN, N., SKEEN, D., AND RIES, D., "DDM: An Ada Compatible Distributed Database Manager," IEEE COMPCON Digests of Papers, (1983).Google Scholar
- 5. DAVIDSON, S. B., GARCIA-MOLINA, HECTOR, AND SKEEN, DALE, "Consistency in Partitioned Networks," ACM Computing Surveys, pp. 341-370 (September 1985). Google ScholarDigital Library
- 6. DEWITT, D., KATZ, R., OLKEN, F., SHAPIRO, L., STONEBRAKER, M., AND WOOD, D., "Implementation Techniques for Main Memory Database Systems," ACM SIGMOD Conference Proceedings, (1984). Google ScholarDigital Library
- 7. FISCHER, M. J. AND MICHAEL, A., "Sacrificing Serializability to Attain High Availability in an Unreliable Network," Proc. of the First ACM SIGACT-SIGMOD Symposium on Principles of Database Systems, (1982). Google ScholarDigital Library
- 8. GARCIA-MOLINA, H., "Performance Comparison of Two Update Algorithms For Distributed Databases," Proc. of 3rd Berkeley Workshop on Distributed Data Management and Computer Networks, pp. 108-119 (Aug. 1978).Google Scholar
- 9. GARCIA-MOLINA, H., "Using Semantic Knowledge for Transaction Processing in a Distributed Database," ACM Trans. on Database Systems, pp. 186-213 (June 1983). Google ScholarDigital Library
- 10. HAGMANN, R., "A Crash Recovery Scheme for a Memory Resident Database System," IEEE Transactions of Computers, pp. 839-843 (September 1986). Google ScholarDigital Library
- 11. HAMMER, M. AND SHIPMAN, D., "Reliability Mechanism for SDD-I: A System for Distributed Databases," ACM Trans. on database Systems, pp. 431-466 (December 1980). Google ScholarDigital Library
- 12. LAMPORT, L., "Time, Clocks and Ordering of Events in Distributed Systems," Communications of ACM, pp. 558-565 (July 1978). Google ScholarDigital Library
- 13. LEHMAN, TOBIN AND CAREY, MICHAEL, "Query Processing in Main Memory Data-base Management Systems," in the Proc. of ACM SIGMOD, pp. 239-250 (1986). Google ScholarDigital Library
- 14. MILENKOVIC, M., "Synchronization of Concurrent Updates in Redundant Distributed Databases," Distributed Data Bases, pp. 49-65 North-Holland Publishing Co., (1980).Google Scholar
- 15. PAPADIMITRIOU, C. H., "Serializability of Concurrent Updates," Journal of ACM, pp. 631-653 (Oct. 1979). Google ScholarDigital Library
- 16. SALEM, K. AND GARCIA-MOLINA, H., "Crash Recovery Mechanisms for Main Storage Database Systems," Tech. Rep. CS-TR-034-86, Dept. of Computer Science, Princeton University, Princeton, (April 1986).Google Scholar
- 17. SHAPIRO, L., "Join Processing in Database Systems with Large Memories," ACM Trans. on Database Systems, pp. 239-264 (September 1986). Google ScholarDigital Library
- 18. SINGHAL, MUKESH, "Concurrency Control Algorithms and Their Performance in Replicated Database Systems," Ph.D. dissertation, Dept. of Computer Science, University of Maryland, College Park, (February, 1986). Google ScholarDigital Library
- 19. SINGHAL, MUKESH AND AGRAWALA, A. K., "A Concurrency Control Algorithm and its Performance for Replicated Database Systems," Proc. of the 6th International Conference on Distributed Computing Systems, Cambridge, Massachusetts, (May 19-23, 1986).Google Scholar
- 20. SON, S. H., "Using Replication for High Performance Database Support in Distributed Real-Time Systems," in the Proc. of the Symposium on Real-Time Systems, pp. 79-86 (December 1987).Google Scholar
- 21. SON, S. H., "A Recovery Scheme for Database Systems with Large Main Memory," in the Proc. of the 11th Annual International Computer Software and Applications Conference, Tokyo, Japan, pp. 422-427 (October 7-9, 1987).Google Scholar
Index Terms
- Issues and approaches to design of real-time database systems
Recommendations
Real-time database systems: present and future
RTCSA '95: Proceedings of the 2nd International Workshop on Real-Time Computing Systems and ApplicationsThe design and implementation of real-time database systems presents many new and challenging problems. Compared with traditional databases, real-time database systems have a distinct feature: they must maintain temporally coherent data while satisfy ...
Supporting predictability in real-time database systems
RTAS '96: Proceedings of the 2nd IEEE Real-Time Technology and Applications Symposium (RTAS '96)Real-time database systems (RTDBSs) have timing constraints in their specifications, such as read times, deadlines and other temporal constraints. In addition, RTDBSs must adapt to changes in the operating environment and guarantee the completion of ...
Comments