Maxime Chevalier-Boisvert introduces adaptive compilation strategies aiming to fulfill the seemingly conflicting goals of producing more optimal machine code while reducing compilation time.
Eddy Bruel details how objects are implemented internally in SpiderMonkey as shape trees and slot vectors. This is done in order to minimize the memory footprint and obtain better JIT optimization with inline caching.
Limin Fu introduces Dao, a lightweight and optionally typed programming language having a LLVM-based JIT compiler optimized for numeric computation, and a Clang-based tool generating Dao bindings for C/C++ libraries.
Charlie Hunt explains what can be done to lower the latency introduced by the Java GC and JIT, including coding tips, and introducing tools for tuning the performance of Java applications.