flow_graph thread harmonization with OpenMP ?

[nyue]

I am developing a node which calls external code/function that are OpenMP aware, what is the best practice of passing the maximum number of threads to the underlying OpenMP code so that everyone play nice and respect the overall threads count availability. Is there some function I can query in flow_graph in regards to available thread so that I don't oversubscribe for the underlying OpenMP code.

Kind regards

[NoorjahanSk_Intel]

Hi,

Thanks for posting in intel communities.

You can use the async_node of OneTBB to call external code/function.

Please refer to the below link for more details:

https://spec.oneapi.io/versions/1.1-rev-1/elements/oneTBB/source/flow_graph/async_node_cls.html

OneTBB workload is balanced dynamically between worker threads every time. The number of worker threads processing a parallel construction always changesdepending on workload. Both OneTbb and OpenMP threads will be active.

Please refer to the below link for more details:

https://oneapi-src.github.io/oneTBB/main/tbb_userguide/appendix_B.html

You can query the number of threads by referring to the below link

https://oneapi-src.github.io/oneTBB/main/tbb_userguide/Migration_Guide/Task_Scheduler_Init.html

Thanks & Regards,

Noorjahan.

[NoorjahanSk_Intel]

Hi,

We haven't heard back from you. Could you please provide an update on your issue?

Thanks & Regards,

Noorjahan.

[NoorjahanSk_Intel]

Hi,

I have not heard back from you, so I will close this inquiry now. If you need further assistance, please post a new question.

Thanks & Regards,

Noorjahan.