After all these years, the task of tuning Kubernetes microservice applications is a daunting task even for experienced Performance Engineers and SREs, often resulting in companies facing reliability and performance issues, as well as unexpected costs.
In this session, we plan to first illustrate some less-known facts about Kubernetes key resource management and autoscaling mechanisms and show how properly setting pod resources and autoscaling policies is critical to avoid over-provisioning while ensuring services deliver the expected performance and resilience.
We then demonstrate how a new approach leveraging ML techniques makes it possible to automatically tune both pod and runtime configurations to ensure any specified optimization goal, such as minimizing Kubernetes cost or maximizing application throughput, while respecting any SLOs, such as max response time and error rates. Results of real-world cases will be used to document how much this new approach can be effective to deliver higher operational efficiency tangible benefits.
Priority access to all content
Community Discord
Exclusive promotions and giveaways