[SYSTEMDS-3704] Resource-aware Linearizers #2197
Open
Conversation
This file contains 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
This commit adds a new linearizer. It is still work in progress. The focus of the new linearizer is on reducing the maximal memory usage. This commit implements the discovery functionality of the new linearizer. The scheduling is work in progress.
This commit adds a new linearizer. The focus of the new linearizer is on reducing the maximal memory usage by finding the permutation with the lowest memory requirement.
This commit finishes the WIP resource aware linearizer.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
This pull request adds two new linearizers for the LOP DAG linearization.
The linearizers focus on reducing the minimal memory requirement by scheduling the LOPs in a way that affects the memory requirement for live intermediates.
RESOURCE_AWARE_OPTIMAL uses a brute-force approach that generates the optimal sequence but requires a lot of time to compile which makes it impractical for larger algorithms and most real world use cases.
RESOURCE_AWARE_FAST uses a heuristic approach that does not always generate the optimal sequence but is a lot faster than the other approach. The linearizer can be used for tasks with large data sets, but there are also some cases where the compile time is larger than expected or the memory requirement is larger in comparison to other linearizer.
cc @mboehm7