Parallel algorithms#120
Open
diehlpk wants to merge 18 commits intocplusplus:masterfrom
Open
Conversation
Discussion with the US group
Mention unit testing
Move building a lib and link it to a C++ executable to the main level
Contributor
Author
|
Co-authored-by: Victor Eijkhout eijkhout@tacc.utexas.edu |
vulder
requested changes
Apr 19, 2026
| @@ -0,0 +1,124 @@ | |||
| ## Module name: Asynchronous programming | |||
Member
There was a problem hiding this comment.
It seems that this PR is combining two topic introductions, could you please replace the asynch one here with a dummy so that your other PR can properly introduce that topic?
| #### Points to cover | ||
|
|
||
| * The header `<execution>` needs to be included | ||
| * The first argument of the algorithm is the execution policy |
Member
There was a problem hiding this comment.
The execution policies them selves and their semantics should also be covered.
| _These are important topics that are not expected to be covered but provide | ||
| guidance where one can continue to investigate this topic in more depth._ | ||
|
|
||
| * If the implementation cannot parallelize or vectorize (e.g. due to lack of resources), all standard execution policies can fall back to sequential execution. |
Member
There was a problem hiding this comment.
This sounds more like a points to cover that should be added above as a student should be aware of that when using the policies.
| guidance where one can continue to investigate this topic in more depth._ | ||
|
|
||
| * If the implementation cannot parallelize or vectorize (e.g. due to lack of resources), all standard execution policies can fall back to sequential execution. | ||
| * None of the execution policies allow for reproducibilty. This is obvious for the parallel execution policies. But even `std::ececution::seq` can execute the iterations in any order. |
Member
There was a problem hiding this comment.
This could also be a caveat or points to cover in main.
Comment on lines
+138
to
+139
| * Nvidia supports to run `std::execution::par` on Nvidia GPUs. However, that is not yet in the C++ standard and will only work with Nvidia's HPC compiler. | ||
| * Currently, parallel algorithms are implemented using Intel's TBB library in GCC. You can set the number of used cores using `tbb::global_control(tbb::global_control::max_allowed_parallelism, nthreads);` provided by the header `#include "tbb/tbb.h"`. |
Member
There was a problem hiding this comment.
This seems to me more like additional information, compared to advanced features about the topic. I'm not sure where to best put this.
Co-authored-by: Florian Sattler <vuld3r@gmail.com>
Co-authored-by: Florian Sattler <vuld3r@gmail.com>
Co-authored-by: Florian Sattler <vuld3r@gmail.com>
Co-authored-by: Florian Sattler <vuld3r@gmail.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Attempt for parallel algorithms by @diehlpk and @VictorEijkhout