Is 40000 cuda streams an issue?

I have a tensor-rt function that is creating 5 streams every time i call it.
Over 200 seconds, it ended up creating 40,000 cuda streams.

Is this an issue performance/resource wise?

Thank you,
Sujan

Hi,
The below link might be useful for you
https://docs.nvidia.com/deeplearning/tensorrt/best-practices/index.html#thread-safety
https://docs.nvidia.com/cuda/cuda-c-programming-guide/index.html#stream-priorities
https://docs.nvidia.com/cuda/cuda-runtime-api/group__CUDART__STREAM.html
For multi threading/streaming, will suggest you to use Deepstream or TRITON
For more details, we recommend you to raise the query to the Deepstream or TRITON forum.

Thanks!

Thanks for the above links but my question is more general.

Is there any limits to the number of cuda streams an application can create?

Thanks.

Hi,

We recommend you please post your query on CUDA related forum to get better help.

Thank you.