Parallel JavaScript

on YouTube ↗23m

" summary

Parallel JavaScript exposes mapPar/reducePar and side‑effect‑free elemental functions and compiles through IonMonkey to CPU or GPU backends, producing native code or OpenCL kernels. Implementation in Firefox Nightly uses static safety analysis plus runtime guards, a work‑stealing scheduler and a GC tuned to parallel runs. Benchmarks claim 8× speedup (dense matrix), 2.8× (2D‑conv), 6.2× (Mandelbrot) and ~7.7× lower energy; GPU code generation targets OpenCL.

AI-written summary · about

" chapters (13)

1. 0:00Start
2. 0:23Parallel Javascript
3. 1:16Why Do We Want To Invent a New Parallel Programming Model
4. 3:44Overview of the Parallel Javascript Api
5. 4:29Elemental Functions
6. 5:23Code Examples
7. 8:17Reversing an Array
8. 10:25Jit Support
9. 12:26Parallel Javascript on Gpus
10. 12:54Why Gpus Typically Have a Separate Address Space than the Cpu
11. 16:45Parallel Safety Analysis
12. 17:06Gpu Specific Safety Analysis
13. 22:07Implementation

" related talks

• 41mConcurrency and Parallel Computing in JavaScriptStephan Herhut · Strange Loop 2013
• 33mParallelism experiments in JavaScriptNaveed Ihsanullah · JSConf US 2015
• 24mShared Memory & Parallel Programming with JavaScriptNidin Vinayakan · JSConf EU 2017
• 25mGetting Over Moore's Law: Parallelization Using JavaScript in the BrowserDamian Schenkelman · JSConf Colombia 2015
• 52mJavaScript at the Speed of LightDmitry Lomov · The Rolling Scopes Conference 2015
• 31mJavaScript Compilation Techniques or "Wasn't JavaScript supposed to be slow?"Marijn Haverbeke · JSConf US 2011
• 32mGPU.JS - GPU Accelerated JavaScriptFazli Sapuan, Eugene Cheah · JSConf.Asia 2016
• 46m10 secrets to improve Javascript PerformanceErick Wendel · JS Fest 2019 Autumn