An history of NVidia Stream Multiprocessor

▤ Summary

Thanks to the abundant documentation published over the years by the green team, I was able to go back in time and fast forward though the fascinating evolution of their stream multiprocessors. The Stream Multiprocessor (SM) replaces all previous units thanks to is ability to run vertex, fragment and geometry "kernel" without distinction. From a programming perspective, the unified memory system of Fermi allowed CUDA C to be augmented with C++ features such as object, virtual methods, and exceptions. This approach made the scheduling logic more complicated (we will revisit this) but with up to six warp instruction per clock a SMX provided 2x the performance of a Fermi SM. Since Intel proved that there is still room for miniaturization with the 7nm of Ice Lake, there is little doubt Nvidia will leverage it to shrink its SM even more and double performance again.

▦ Linked Companies

NVIDIA Corporation

$1B+

TechPowerUp

$1M to $5M