Benchmarking Distributed Stream Data Processing Systems

Benchmarking Distributed Stream Data Processing Systems

Karimov, Jeyhun (German Research Centre for Artificial Intelligence (DFKI))
Rabl, Tilmann (Technical University of Berlin; German Research Centre for Artificial Intelligence (DFKI))
Katsifodimos, A (TU Delft Web Information Systems)
Samarev, Roman (German Research Centre for Artificial Intelligence (DFKI))
Heiskanen, Henri (Rovio Entertainment)
Markl, Volker (German Research Centre for Artificial Intelligence (DFKI); Technical University of Berlin)

2018-10-24

The need for scalable and efficient stream analysis has led to the development of many open-source streaming data processing systems (SDPSs) with highly diverging capabilities and performance characteristics. While first initiatives try to compare the systems for simple workloads, there is a clear gap of detailed analyses of the systems' performance characteristics. In this paper, we propose a framework for benchmarking distributed stream processing engines. We use our suite to evaluate the performance of three widely used SDPSs in detail, namely Apache Storm, Apache Spark, and Apache Flink. Our evaluation focuses in particular on measuring the throughput and latency of windowed operations, which are the basic type of operations in stream analytics. For this benchmark, we design workloads based on real-life, industrial use-cases inspired by the online gaming industry. The contribution of our work is threefold. First, we give a definition of latency and throughput for stateful operators. Second, we carefully separate the system under test and driver, in order to correctly represent the open world model of typical stream processing deployments and can, therefore, measure system performance under realistic conditions. Third, we build the first benchmarking framework to define and test the sustainable performance of streaming systems. Our detailed evaluation highlights the individual characteristics and use-cases of each system.

Apache Flink
Apache Spark
Apache Storm
Stream benchmark
Stream data processing

http://resolver.tudelft.nl/uuid:427b434d-6e9b-4fd9-81d8-49a265cd90ac

https://doi.org/10.1109/ICDE.2018.00169

IEEE, Piscataway, NJ

978-1-5386-5520-7

Proceedings - IEEE 34th International Conference on Data Engineering, ICDE 2018

34th IEEE International Conference on Data Engineering, ICDE 2018, 2018-04-16 → 2018-04-19, Paris, France

Institutional Repository

conference paper

© 2018 Jeyhun Karimov, Tilmann Rabl, A Katsifodimos, Roman Samarev, Henri Heiskanen, Volker Markl