California-based startup REELY and a member of the NVIDIA Inception Program developed a deep learning platform that automatically generates sports highlights in real-time.
A defining feature of the new Volta GPU Architecture is its Tensor Cores, which give the Tesla V100 accelerator a peak throughput 12 times the 32-bit floating point throughput of the previous-generation Tesla P100.
In this post we introduce the “register cache”, an optimization technique that develops a virtual caching layer for threads in a single warp. It is a software abstraction implemented on top of the NVIDIA GPU shuffle primitive.
