How can we utilize GPU shared memory?

2 views (last 30 days)
MathWorks Support Team
MathWorks Support Team on 30 Jul 2018
Commented: Walter Roberson on 25 Feb 2019
How can we utilize GPU shared memory? I got the error "GPU out of memory". However, the task manager said there was a shared GPU memory of 7.9 GB which was not utilized.

Sign in to answer this question.

Accepted Answer

MathWorks Support Team
MathWorks Support Team on 30 Jul 2018
The "shared GPU memory" shown in task manager is not memory on the GPU itself, but part of the overall system memory that Windows makes available to the GPU. Unfortunately, we do not support utilizing that memory.
  1 Comment
Walter Roberson
Walter Roberson on 25 Feb 2019
No, it will not be supported in near future. It is forced by MS Windows, and NVIDIA and Microsoft will have to negotiate to solve the problem.
It is not clear what list you are referring to about bottlenecks and sparse matrix support ?

Sign in to comment.

More Answers (0)

Categories

Find more on GPU Computing in MATLAB in Help Center and File Exchange

Tags

Products

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!