Don't have Telegram yet? Try it now!
https://hgpu.org/?p=8209
Reducing thread divergence in a GPU-accelerated branch-and-bound algorithm