Fast and Compact - Reducing Size of AOT-Compiled Java Code Without Sacrificing Performance
ACM SIGPLAN via YouTube
AI Engineer - Learn how to integrate AI into software applications
AI, Data Science & Cloud Certificates from Google, IBM & Meta
Overview
AI, Data Science & Cloud Certificates from Google, IBM & Meta — 40% Off
One plan covers every Professional Certificate on Coursera. 40% off Coursera Plus Annual.
Unlock All Certificates
Explore a conference presentation that introduces an innovative approach to reducing the size of GraalVM Native Image executables while maintaining high performance. Learn about a novel technique that selectively compiles frequently used methods to machine code while keeping rarely executed methods as bytecode, executed through a lightweight interpreter. Discover how this hybrid approach can simulate JIT compilation behavior when interpreted methods become performance-critical, achieving up to 36.9% reduction in executable size for DaCapo benchmark applications without compromising peak performance or startup times. Understand the technical implementation details of this AOT compilation optimization strategy and its implications for Java application deployment in resource-constrained environments.
Syllabus
[MPLR'25] Fast and Compact: Reducing Size of AOT-Compiled Java Code Without Sacrificing Performance
Taught by
ACM SIGPLAN