DirectML and GPU Shared memory #19395
NarutoUA
started this conversation in
Ideas / Feature Requests
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Once onnxruntime-dml fully utilizes dedicated VRAM, it resorts to utilizing shared system memory, leading to a significant decrease in performance. I suggest introducing an execution provider option to toggle the utilization of system memory
Beta Was this translation helpful? Give feedback.
All reactions