Evaluating YJIT's Performance in a Production Context - A Pragmatic Approach
Offered By: ACM SIGPLAN via YouTube
Course Description
Overview
Explore a 20-minute conference talk from MPLR 2023 that evaluates the performance of YJIT, a new JIT compiler for Ruby, in a production context. Discover how YJIT, based on Lazy Basic Block Versioning architecture, compares to other Ruby implementations in terms of peak performance, memory usage, and warm-up time. Learn about YJIT's consistent outperformance of the CRuby interpreter on benchmarks and its significant speedups in real-world deployments, such as Shopify's StoreFront Renderer infrastructure. Gain insights into YJIT's advantages, including full compatibility with existing Ruby code, reduced memory overhead, and faster warm-up compared to JRuby and TruffleRuby. Understand the pragmatic approach taken to evaluate YJIT's performance in both benchmark and large-scale production environments.
Syllabus
[MPLR23] Evaluating YJIT’s Performance in a Production Context: A Pragmatic Approach
Taught by
ACM SIGPLAN
Related Courses
Web Application ArchitecturesUniversity of New Mexico via Coursera Ruby مدخل إلى برمجة مواقع الإنترنت باستخدام لغة
Rwaq (رواق) Rails with Active Record and Action Pack
Johns Hopkins University via Coursera Ruby on Rails: An Introduction
Johns Hopkins University via Coursera Ruby on Rails Web Services and Integration with MongoDB
Johns Hopkins University via Coursera