The impact of workload on simulation results for distributed transaction processing
To use benchmarks or to use real database traces (for the simulation of high performance distributed transaction processing), that is the question. Whether it is preferable to deal with special real workload, or to ignore the rich diversity observable with real measurements by restricting oneself to standardized benchmarks. We demonstrate how the evaluation and comparison of different load balancing algorithms depends on the workload considered. For that purpose we compare the TPC-B benchmark with two real workload traces: We analyze their properties with respect to workload characteristics on a theoretical level and we indicate how the differences observed influence performance evaluations in simulation experiments. The results imply that performance of different load distribution algorithms essentially depends on the workload processed.
KeywordsReal Trace Page Access Workload Characteristic Real Workload Page Buffer
Unable to display preview. Download preview PDF.
- 1.N. Auerbach: Lastclusterung für verteilte Hochleistungsdatenbanksysteme, Diplomarbeit, University of Zurich, 1997.Google Scholar
- 2.E. Born, T. Delica, W. Ehrl, L. Richter, R. Riedl: Characterization of Workloads for Distributed DB/DC-Processing, Journal on Information Science, 1997Google Scholar
- 3.R. Dubes, A. Jain: Algorithms for Clustering Data, Prentice-Hall, 1988Google Scholar
- 4.S. Haldar and D. Subrammanian: An affinity-based dynamic load balancing protocol for distributed transaction processing systems, Performance Evaluation 17, 1993Google Scholar
- 5.Load Balancing in High-Performance Distributed DB/DC-Systems: Modeling, Analysis and Clustering of Workload for Data-Affinity Based Load Distribution, ALV Workshop Munich, 1998Google Scholar
- 6.R. Riedl and R. Salomon; Proc. 2nd ICSC Symp. Fuzzy Control of Load-Balancing for Distributed DB/DC Transaction Processing, Fuzzy Logic and Applications, 1997Google Scholar
- 7.W. Shakespeare; The Tragedy of Hamlet, Prince of Denmark, 3.1Google Scholar
- 8.Yu P.S., Cornell C., Dias M., Balakrishna I.: Performance Analysis of Affinity Clustering on Transaction Processing Coupling Architectures, IEEE Trans. Knowledge and Data Engineering, Vol 6, No 5, 1994Google Scholar
- 9.http://www.ics.forth.gr/pleiades/projects/LYDIA/Google Scholar