A Performance Study of Event Processing Systems

Authors

Abstract

Event processing engines are used in diverse mission-critical scena-rios such as fraud detection, traffic monitoring, or intensive care units. Howev-er, these scenarios have very different operational requirements in terms of, e.g., types of events, queries/patterns complexity, throughput, latency and num-ber of sources and sinks. What are the performance bottlenecks? Will perfor-mance degrade gracefully with increasing loads? In this paper we make a first attempt to answer these questions by running several micro-benchmarks on three different engines, while we vary query parameters like window size, win-dow expiration type, predicate selectivity, and data values. We also perform some experiments to assess engines scalability with respect to number of que-ries and propose ways for evaluating their ability in adapting to changes in load conditions. Lastly, we show that similar queries have widely different perfor-mances on the same or different engines and that no engine dominates the other two in all scenarios.

Subject

Complex Event Processing