Benchmarking Distributed Stream Data Processing Systems

TL;DR

This paper introduces a benchmarking framework for distributed stream processing systems, evaluating throughput and latency of windowed operations on real-world workloads, providing detailed performance insights into Apache Storm, Spark, and Flink.

Contribution

It defines latency and throughput for stateful operators, separates system and driver for realistic testing, and creates the first comprehensive benchmarking framework for streaming systems.

Findings

Flink outperforms in throughput and latency for windowed operations.

System performance varies significantly across different workloads.

Benchmarking framework reveals each system's unique strengths and limitations.

Abstract

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.