Skip to content
Open access

Benchmarking Resilience of Stream Processing Middleware Deployed on the Edge

Oct 2026 · Kiel Software Engineering Research · 0 citations

Abstract

Shifting centralized server architectures to edge devices enhances privacy and reduces network bandwidth consumption. To enable this transition, stream processing middleware on edge devices is required to pre-process and manage data effectively. In this work, the cloud-native, distributed event streaming platform Apache Kafka is deployed on simulated edge infrastructure and subjected to controlled fault scenarios using Chaos Mesh. These fault scenarios map typical edge-related resource constraints to determine write-path’s operational suitability and derive practical deployment guidelines for the edge. The empirical results show that Kafka’s write-path can handle the evaluated streaming workload on resource-constrained, simulated edge devices. Across all fault scenarios, applied chaos significantly decreases incoming data. While CPU stress increases internal queue times drastically (exceeding 1000%), added network latency causes pipeline starvation, depletes queues, and increases remote time, likely due to replication. The combined stress scenario reveals an interplay between the opposing queue dynamics, resulting in intermediate queue behaviour. Based on these insights, deployment guidelines for edge middleware operators are presented.

Read PDF

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.