Maximizing Performance with Massively Parallel Hash Maps on GPUs

Originally published at: https://developer.nvidia.com/blog/maximizing-performance-with-massively-parallel-hash-maps-on-gpus/

Learn the fundamentals of hash maps and how their memory access patterns make them well suited for GPU acceleration.

2 Likes

Thank you very much for this awesome article on using hash maps from cuCollections to accelerate lookups. I was wondering if the source code for making a Tensorflow custom op using cuCollections hash maps is available, as it seems that several companies (such as Pinterest) use it.