search

mratsim / weave

A state-of-the-art multithreading runtime: message-passing based, fast, scalable, ultra-low overhead
Other
541 stars 21 forks source link

40+% overhead reduction of memory pool on overhead-bound benchmarks #113

Closed mratsim closed 4 years ago

mratsim commented 4 years ago

This magic 1-liner commit reduces the overhead of eager futures by over 40% and closes #112

image

We are back to being way under Intel TBB (600 ms) and Intel TBB has the lowest overhead compared to all frameworks used in production for task parallelism: OpenMP (GCC, LLVM/Intel) or Julia Partr or HPX image