How to display latency, memory ops, and arithmetic ops in Compute Profiler

Accelerated Computing CUDA CUDA Programming and Performance
1

Hey all,

I heard at the GTC Conference that with the compute profiler, it should be possible to get a comparison of how much time is being spent for arithmetic ops, memory ops, or wasted on latency. I searched the profiler after running my program and I tried googling, but I don’t see anything related to figuring out this metrics.

Can anybody help, is my question clear?

2

Hmmm this is weird that nobody has responded. Nobody responded to a similar question I posed on Stack Overflow. This latency vs memory ops vs instruction ops profiling was prominently displayed at the Nvidia GTC Conference. I am wondering if it is completely new and or not released yet. If anybody has any ideas let me know, I am going to try and find the name of the presenter and Email him. He did the last lecture of Wed and Thurs at the conference.

3

Hmmm this is weird that nobody has responded. Nobody responded to a similar question I posed on Stack Overflow. This latency vs memory ops vs instruction ops profiling was prominently displayed at the Nvidia GTC Conference. I am wondering if it is completely new and or not released yet. If anybody has any ideas let me know, I am going to try and find the name of the presenter and Email him. He did the last lecture of Wed and Thurs at the conference.