Boosting Microservice Resilience: An Evaluation of Istio’s Impact on Kubernetes Clusters Under Chaos

Microservices based architecture is increasingly popular for deploying cloud-native applications using small, decoupled modules. While this approach provides flexibility, it also increases the complexity to manage and maintain the application. Microservices are deployed on Kubernetes, which enables efficient resource management and scalability. However, resilience depends on individual services and pods. Chaos engineering tests resilience by injecting failures. Istio service mesh can improve stability and fault tolerance when deployed as part of the microservice. This paper explores the use of chaotic tests to evaluate Istio’s service’s impact on microservices resilience in Kubernetes, and optimal tuning. The performance is evaluated under increased load, measuring and analysing metrics such as response time, errors, resource usage, and requests per second. Results show the Istio service mesh offers better stability, faster responses, and failure resilience compared to traditional architectures.