Ruby3.0 and Beyond - Yukihiro "Matz" Matsumoto
summary
Ruby 3.0 ships December 25 2020 with concrete claims: a goal of 3x faster claim versus Ruby 2.0 backed by MJIT/“jet” compilation and microbenchmark results (NES emulator 32.5 → 98.4 FPS with JIT). Concurrency gains come from fibers for async IO and Ractor experimental isolated actors showing ~3.6× parallel speedups in a tarai benchmark. Static tooling includes RBS and TypeProf for generated signatures; MJIT tradeoffs (compile cost and hosting limits) and ongoing performance work are detailed.
AI-written summary · about
chapters (16)
- 0:00Introduction
- 1:23Ruby 3.0 release goals
- 3:09Compatibility vs progress
- 7:13Ruby 1.9 lessons
- 11:29Why compatibility matters
- 13:20Performance in Ruby 3.0
- 17:57Ruby 3.0 performance
- 23:03MJIT challenges
- 26:16Concurrency with fibers
- 34:36Static analysis RBS TypeProf
- 41:21Pattern matching new syntax
- 43:54Ruby future tooling
- 46:29Performance roadmap JIT
- 49:21Static barriers dynamic Ruby
- 51:12Closing remarks
- 51:37Q&A
