ShuffleBench: A Benchmark for Large-Scale Data Shuffling Operations with Distributed Stream Processing Frameworks

TL;DR

Results show that Flink achieves the highest throughput while Hazelcast processes data streams with the lowest latency, and an experimental performance evaluation that employs ShuffleBench with various configurations on Flink, Hazelcast, Kafka Streams, and Spark in a cloud-native environment.

Abstract

Distributed stream processing frameworks help building scalable and reliable applications that perform transformations and aggregations on continuous data streams. This paper introduces ShuffleBench, a novel benchmark to evaluate the performance of modern stream processing frameworks. In contrast to other benchmarks, it focuses on use cases where stream processing frameworks are mainly employed for shuffling (i.e., re-distributing) data records to perform state-local aggregations, while the actual aggregation logic is considered as black-box software components. ShuffleBench is inspired by requirements for near real-time analytics of a large cloud observability platform and takes up benchmarking metrics and methods for latency, throughput, and scalability established in the performance engineering research community. Although inspired by a real-world observability use case, it is highly configurable to allow domain-independent evaluations. ShuffleBench comes as a ready-to-use open-source software utilizing existing Kubernetes tooling and providing implementations for four state-of-the-art frameworks. Therefore, we expect ShuffleBench to be a valuable contribution to both industrial practitioners building stream processing applications and researchers working on new stream processing approaches. We complement this paper with an experimental performance evaluation that employs ShuffleBench with various configurations on Flink, Hazelcast, Kafka Streams, and Spark in a cloud-native environment. Our results show that Flink achieves the highest throughput while Hazelcast processes data streams with the lowest latency.

Figures (8)

• Figure 1: Overview of ShuffleBench benchmark components according to the Empirical Standard for benchmarking.
• Figure 2: The ShuffleBench dataflow architecture at runtime for three stream processing application instances.
• Figure 3: Baseline throughput results obtained with the ad-hoc measurement method of Flink, Hazelcast, and Kafka Streams (a); of Spark for different limitations on the maximum number of pulled records per batch (b); and with the sustainable measurement method of Flink, Hazelcast, Kafka Streams, and Spark (c).
• Figure 4: Quantile function of the median observed latency of Flink, Hazelcast and Kafka Streams.
• Figure 5: Experimental results comparing a deployment with 9 instances and one core per instance with a deployment of 3 instances with 3 cores per deployment.