OCaml Forge

Forum: high-performance-parallelism-with-hlvm

Posted by: Jon Harrop
Date: 2010-01-03 18:30
Summary:High-performance parallelism with HLVM

The first-working version of parallelism in HLVM was very inefficient due to repeated fetching of thread-local data via pthreads. The latest update passes thread-local data to every function in an auxiliary argument and achieves production-level performance: 2× faster than OCaml on average over non-list benchmarks.

Latest News

Variant types and pattern matching in HLVM

Jon Harrop - 2010-04-28 09:37 -

HLVM on the ray tracer language comparison

Jon Harrop - 2010-01-21 17:01 -

High-performance parallelism with HLVM

Jon Harrop - 2010-01-03 18:30 -

Parallel programming in HLVM

Jon Harrop - 2009-11-30 09:08 -

New HLVM GC and LLVM 2.6 support

Jon Harrop - 2009-10-28 23:32 -

Discussion Forums: high-performance-parallelism-with-hlvm

Monitor Forum | Start New Thread Start New Thread