A measure of transaction processing power
Datamation
Benchmarking Database Systems A Systematic Approach
VLDB '83 Proceedings of the 9th International Conference on Very Large Data Bases
ACM SIGMOD Record - Tribute to honor Jim Gray
Hi-index | 0.01 |
Today's DBMS are constantly upon large-scale workloads (e.g., internet) and require a reliable tool to benchmark them upon a similar workload. Usually, benchmarking tools simulate a multi-user workload within a single machine. However, this avoids large-scale benchmarking and also introduces deviations in the result. In this paper, we present a solution for benchmarking DBMS in a fully distributed manner. The solution is based on the TPC-C specification and aims to simulate large-scale workloads. We validate our solution through implementation and experimentation on three open-source DBMS. Through experimentation, we analyze the behavior of all systems and show how our solution is able to scale up the benchmark.