Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Disable TBB by default #1762

Merged
merged 1 commit into from
Aug 11, 2020
Merged

Disable TBB by default #1762

merged 1 commit into from
Aug 11, 2020

Conversation

joe-maley
Copy link
Contributor

@joe-maley joe-maley commented Aug 10, 2020

This disables TBB by default in the bootstrap script.

Using our /test/benchmarking benchmarks, there is no discernible difference
between the end-to-end execution times or the individual execution times of the
parallel functions. For example, here is are the results of the bench_large_io
between the two:

// Without TBB
end-to-end: 267829 ms
parallel_for: 271120 ms
parallel_for_2d: 104 ms
parallel_sort: 14447 ms

// With TBB
end-to-end: 269911 ms
parallel_for: 273447 ms
parallel_for_2d: 107 ms
parallel_sort: 14382 ms

Note that the "end-to-end" was measured with a different timing mechanism than
the individual function timers. The individual function timers are also
cumulative per thread.

@joe-maley joe-maley changed the title Jpm/disable tbb Disable TBB by default Aug 10, 2020
Copy link
Member

@Shelnutt2 Shelnutt2 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Please update the history file

@joe-maley joe-maley force-pushed the jpm/disable-tbb branch 2 times, most recently from 23f3210 to dd8e8b9 Compare August 10, 2020 22:56
@joe-maley joe-maley requested a review from Shelnutt2 August 11, 2020 10:50
This disables TBB by default in the bootstrap script.

Using our `/test/benchmarking` benchmarks, there is no discernible difference
between the end-to-end execution times or the individual execution times of the
parallel functions. For example, here is are the results of the `bench_large_io`
between the two:

// Without TBB
end-to-end: 267829 ms
`parallel_for`: 271120 ms
`parallel_for_2d`: 104 ms
`parallel_sort`: 14447 ms

// With TBB
end-to-end: 269911 ms
`parallel_for`: 273447 ms
`parallel_for_2d`: 107 ms
`parallel_sort`: 14382 ms

Note that the "end-to-end" was measured with a different timing mechanism than
the individual function timers. The individual function timers are also
cumulative per thread.
@joe-maley joe-maley merged commit 466d5ef into dev Aug 11, 2020
@joe-maley joe-maley deleted the jpm/disable-tbb branch August 11, 2020 13:44
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants