Calibrating JVM on K8s

Developing a sound and efficient application is undeniably the biggest and the most complex part of creating a system. But still the operations part is something that might screw up all the coding effort. Especially, when rare app load peaks are making your system crash or become hardly usable. Of course, you can always negotiate with a manager bigger budget for infrastructure to triple the resources, explaining that highly resilient system costs money, but why not just understand the app resource consumption for good and configure it in effective way?

Talk given by Developer to Developers who are willing to broaden their scope of skills and responsibility and become more DevOps-like.
It will go from the very bottom to high top, starting with low details on how OS process allocates memory and how it's given CPU time, then how JVM relies on these system basics, up to how virtual machine cooperates with Docker containerization environment. All this to show how top level Kubernetes resource request/limit parameters is translated down to lower levels of abstraction.

Theory aspects chosen in pragmatic manner, enriched with live demos of tools that helps to calibrate execution parameters and monitor runtime performance.

Scheduled on Thursday from 14:40 to 15:30 in Room 4