A Performance Benchmark of NetFlow Data Analysis on Distributed Stream Processing Systems
Milan Čermák, Daniel Tovarňák, Martin Laštovička, Pavel Čeleda · Veřejné služby Informačního systému (Masarykiana Brunensis Universitas) · 2016
Modern distributed stream processing systems can be potentially applied to real time network flow processing. However, differences in performance make some systems more suitable than others for being applied in this domain. We propose a novel performance benchmark which is based on a common security analysis algorithms of NetFlow data, to determine the suitability of distributed stream processing system. Three of the most used distributed stream processing systems are benchmarked and results are compared with identified NetFlow data processing challenges and requirements. Benchmark results show that each system reached a sufficient data processing speed using basic deployment scenario with little to no configuration tuning. Our benchmark, unlike any other, enables to determine a performance of processing small structured messages on any stream processing system.