Chapter 8 Analysis of filtering efficiency in the database machine verso
The Relational Database Machine VERSO uses a hardware mechanism with filters data upon its transfer from mass memory to main memory. This paper studies the efficiency of this Filter. A probabilistic modelling approach for evaluating the Filter utilization is introduced. In order to illustrate this approach we perform the analysis for the Union operation under a simple algorithm. The modelling methodology we propose is indeed general and will be used in further studies to analyze the Filter performance on other operations, and in particular Join, which is the most critical one in a relational system.
Unable to display preview. Download preview PDF.
- [BAS80]F.Bancilhon and M.Scholl, "Design of a Backend Procesor for a Database Machine", Proceedings of the ACM-SIGMOD Conference, Santa Monica, California, May 14–16, 1980.Google Scholar
- [BAF80]F.Bancilhon, D.Fortin, S.Gamerman, J.M.Laubin, P.Richard, M.Scholl, D.Tusera and A.Verroust, "A Relational Back-end Database Machine", Proceedings of the International Workshop on Database Machine, San Diego, California, September 1982.Google Scholar
- [ULL80]J.D. Ullman, "Principles of Database Systems”, Computer Science Press, New York, 1980.Google Scholar
- [CIN75]E. Cinlar, "Introduction to Stochastic Processes", Prenticehall, Englewood Cliffs, N.J., 1975.Google Scholar