Conf42 SRE: Let the machines optimize the machines: ML-driven automated performance tuning

Akamas CTO, Stefano Doni will speak at Conf42 SRE online conference on September 30th about how ML-driven automated performance tuning can support SREs in their goal of achieving optimal application performance, efficiency and availability.

During his session, Stefano will first discuss the challenge of configuring today’s complex IT stacks to avoid poor performance and incidents by tuning hundreds of parameters (e.g. JVM and DBMS settings, container CPU and memory) which may interact in ways that are counterintuitive even for the most experienced performance engineers.

The Akamas ML-based approach will be then introduced and applied to a real-world Kubernetes microservices application with the goal of improving cost efficiency and latency by automatically tuning container sizing and JVM options. This example will be used to demonstrate how this approach allows SREs to achieve higher application performance in days instead of months, by simply defining goals and constraints for each specific service (e.g. minimize resource footprint while matching latency and throughput SLOs) and letting ML-based optimization to find the best configuration.