Jan Neumann presents how Comcast uses machine learning and big data processing to facilitate search for users, for capacity planning, and predictive caching.
John Oliver takes a look at both G1 and Shenandoah, explaining how they work, what are their limitations, providing tuning advice. He also looks at recent and future changes to garbage collection.
Monica Beckwith talks about G1 pause (young and mixed) composition, G1's remembered sets and collection set and G1's concurrent marking algorithm, providing performance tuning advice.
Mathieu Bastian explores the mechanics of unit, integration, data and performance testing for large, complex data workflows, along with the tools for Hadoop, Pig and Spark.
Mark Price explores the life cycle of Java code, and how the JVM evolves the runtime representation of code during program execution, providing tips to make sure Java code runs fast.
Ben Hall shares his experience working with Docket for development, testing and deployment into production, discussing scalability, resource management, security and other related issues.
Chris Newland discusses performance-boosting techniques used by the JVM’s JIT and introduces JITWatch, a tool helping to get the best JVM performance for a code.
Jon Moore talks about distributed monotonic clocks (DMC) whose timestamps can reflect causality but which have a component that stays close to wall clock time.
Tony Printezis presents how services are deployed and monitored at Twitter, the benefits of using a custom-built JVM, and the challenges of the use of the JVM in an environment like Twitter.
Brendan Gregg focuses on broken tools and metrics instead of the working ones. Metrics can be misleading, and counters can be counter-intuitive. He advises on how to approach new performance tools.
Cliff Click takes a look at Java vs C performance. He discusses both languages' strong and weak points and the programming context surrounding language choices.
Nitsan Wakart discusses concrete cases in which profilers misguide, misrepresent and at times subvert the systems they aim to help us diagnose.