Due to the complexity of Kubernetes microservices applications, typically many days and weeks of manual effort are spent, just for tuning one single microservice. However, overprovisioning neither avoids slowdowns or resilience issues nor helps make the delivered service cost-effective.
This short video shows how Akamas AI-powered optimization can effectively address these challenges by automatically identifying the optimal configuration of Kubernetes container sizing and JVM runtime parameters, with respect to both the desired performance, resilience or cost-efficiency goals and to the required SLOs.