With the creation of Martin Thompson’s Disruptor and Peter Lawrey’s Chronicle frameworks, Java developers are learning that it is possible to get tremendous throughput in an application on a single thread just by understanding how the hardware tries to optimize your code at runtime. So how do they do it?
One of the biggest challenges facing JVM-based developers is taking advantage of low-level hardware to increase performance in a virtualized platform. Many of us don’t even understand the impact of selecting the right data structure to take advantage of their characteristics with respect to the algorithm we’re attempting to implement. But by understanding how hardware attempts to optimize memory access, you will get a better picture of the impact of your coding choices on application performance.
Join me at JAXConf 2013 where we will discuss the CPU’s memory ecosystem, breaking it down into its most elemental components and how they work. We will then discuss how best to take advantage of them to optimize performance. Join me in at JAXConf in Santa Clara!