The JVM provides us with many ways to implement "mutual exclusion" in multithreaded applications. Synchronized, ReentrantLock, or Atomic. Each of these mechanisms has its pros and cons, also known as "it's just an implementation detail". During the next installment of the journey into the depths of the JVM, you will learn how these mechanisms differ, we will look inside them and you will learn what impact these small implementation details have on performance and scalability. There will also be time for discussions and practical advice on what situations to use each of these mechanisms, and how to trace its behavior.

What is "adaptive locking"? What do threads do when they are parked? What is the mysterious futex and why are Atomic variables are locks at the hardware level? There will time to also talk about reentrance, fairness, and monitor deflation. If this not enough we will close a short introduction to lock-free programming.

If you answered yes to at least one question, or you just want to have fun and not necessarily get anything from this presentation, you are welcome.

Events are great, but they can also bite.

We will talk about what everybody would prefer to avoid — mistakes possible while building and working with events.

Expect: problems, symptoms, consequences, and some 'get-out-of' cards.

Come for the experience others gained first-hand, so you know problems are inbound, and how to avoid them.