Sling memory deep dive

Physical memory is one of the most critical system resources (along with disk IO and CPU) that a Sling application relies on to achieve maximum performance and best response times. Understanding where and how much is used by Sling (with Oak Segment Tar) helps developers optimize their code and it's a required knowhow for devops for sizing, monitoring and debugging live instances.

This session takes a holistic approach to understanding RAM consumption by offering a detailed split-view of all the various flavors of a system's memory: kernel, memory mapped files cache, heap, stack. The focus is laid on guiding the users to setup the system and the application for the best performance. A deep dive into the heap of a running sling application provides even more insights on the most memory-hungry parts of the application for those that are willing to go the extra mile into performance optimization.

Online-Version of the presentation