Intel® oneAPI Threading Building Blocks
Ask questions and share information about adding parallelism to your applications when using this threading library.
Announcements
This community is designed for sharing of public information. Please do not share Intel or third-party confidential information here.

Parallel For-Each

james_b1
Beginner
108 Views
Hiya,

I'm looking to use parallel_for_each to process a concurrent_ordered_map of AI actors.

From the documentation, it looks like a grain-size of 10,000 is a good starting place. The simulation is unlikely to have more than 4000 actors though - so if I understand correctly, only one thread will be used with this value.

Is this right?
0 Kudos
2 Replies
RafSchietekat
Black Belt
108 Views
10000 is often mentioned as the minimal number of processor instructions for a task, not the minimal number of iterations. If that is not clear somewhere, you should probably report it for correction.

But normally you should be able to use auto_partitioner (the default) and not have to specify a grainsize at all (using default 1).
james_b1
Beginner
108 Views
Aha.. Thanks for your quick reply - I misunderstood.
Reply