You do not have permission to delete messages in this group
Copy link
Report message
Show original message
Either email addresses are anonymous for this group or you need the view member email addresses permission to view the original message
to Caffe Users, egor.s...@gmail.com
Hi everyone,
We are using Caffe for inference on a Windows computer in GPU mode. I noted that 1 inference process may process about 24 images per 1 second (GPU load is about 80%, CPU load = 1 fully loaded CPU core). After this I launch another one inference process, GPU load becomes about 85% and after this each process may process about 12 images per 1 second. And it looks absolutely logical for such GPU load - the processes just share GPU power. But at the same time CPU load becomes twice greater than in case of the only process (2 CPU cores becomes fully loaded). And this fact surprises me because in both cases I process 24 images per second, but in the second case it takes 2X more CPU power. Can somebody explain the reason of this behaviour? Or may be it's my error and somebody knows how to deal with it.