Design and evaluation of dynamic optimizations for a Java just-in-time compiler

Abstract

The high performance implementation of Java Virtual Machines (JVM) and Just-In-Time (JIT) compilers is directed toward employing a dynamic compilation system on the basis of online runtime profile information. The trade-off between the compilation overhead and performance benefit is a crucial issue for such a system. This article describes the design and implementation of a dynamic optimization framework in a production-level Java JIT compiler, together with two techniques for profile-directed optimizations: method inlining and code specialization. Our approach is to employ a mixed mode interpreter and a three-level optimizing compiler, supporting level-1 to level-3 optimizations, each of which has a different set of trade-offs between compilation overhead and execution speed. A lightweight sampling profiler operates continuously during the entire period while applications are running to monitor the programs' hot spots. Detailed information on runtime behavior can be collected by dynamically generating instrumentation code that is installed to and uninstalled from the specified recompilation target code. Value profiling with this instrumentation mechanism allows fully automatic profile-directed method inlining and code specialization to be performed on the basis of call site information or specific parameter values at the higher optimization levels. The experimental results show that our approach offers high performance and low compilation overhead in both program startup and steady state measurements in comparison to the previous systems. The two profile-directed optimization techniques contribute significant portions of the improvements.

AOT vs. JIT: impact of profile data on code quality

“…Dynamic or online profiling collects profile information during the same program run, and is commonly employed by advanced managed language run-times, like those for Java [Paleczny et al, 2001, Arnold et al, 2011, Suganuma et al, 2005, Cierniak et al, 2000. Researchers have also developed static analysis techniques to estimate some run-time information for PGOs [Wu & Larus, 1994].…”

Untitled

“…These statistical and probabilistic information can be obtained using edge profiling [15][16][17][18] . In absence of edge profiling, heuristics can be used.…”

A component infrastructure for performance and power modeling of parallel scientific applications

“…A technique known as statistical sampling that is based on profiling mitigates data collection overheads by capturing a subset of events. Statistical sampling involves periodically sampling system counters.Performance tools such as PerfSuite [44], VTune [67], HPCToolkit [38], and JRockit [52] collect performance data using statistical sampling. A drawback of statistical sampling is that the performance data can be less precise and complete compared to tracing.…”