You are using Java, whew!!! No need to worry about memory, the garbage collector will handle that. Those who have had a memory issue in Java are not so naive any more. Often memory utilization and heap sizes are an after thought and are not recognized until the application is in production, often caused by application uptime, production request volume or production sets of data. When the OutOfMemory Error occurs, often the science of development seems to brake down and knobs are turned. First the (-mx) maximum heap space gets adjusted... More is better right. The next OutOfMemory, heads start scratching, code reviews start in earnest, and Google gets several new hits. Did you know that it is possible to get an OutOfMemory error without running out of heap space?

This talk will walk through the underlying details of memory management in the JVM with a focus on VM flags available to help configure the VM. However we can't configure the VM without a detailed understanding of what is going on inside the VM. We'll focus on tools available for analyzing the memory in a running VM. Two actual client case examples will be presented. We'll discuss the differences between the two cases and why the end configurations were quite different.