Dev memo #964
Merged
Dev memo #964
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
Step 1: `AbstractVariable`
…Constraints for the moment)
jgFages
reviewed
Nov 15, 2022
| int nsize = ArrayUtils.newBoundedSize(last, from.length * 2); | ||
| from = Arrays.copyOf(from, nsize); | ||
| to = Arrays.copyOf(to, nsize); | ||
| causes = Arrays.copyOf(causes, nsize); |
There was a problem hiding this comment.
those refactoring seem a very good initiative
jgFages
reviewed
Nov 15, 2022
| @@ -519,7 +522,47 @@ public ICause getCause() { | |||
| return cause; | |||
| } | |||
|
|
|||
| static class BipartiteList { | |||
| static abstract class ABipartiteList { | |||
There was a problem hiding this comment.
Suggestions :
- externalize those classes to reduce file size
- it seems the abstract class has no field and could then be an interface with default implementation (the default implem could also be done as a separate class).
jgFages
reviewed
Nov 15, 2022
| /** | ||
| * Empty bipartite list | ||
| */ | ||
| static final ABipartiteList EMPTY = new ABipartiteList() { |
There was a problem hiding this comment.
in the end the JVM will create an anonymous class, maybe we could have a proper class for the default case
|
Thank you for the review, I took your comments into account and move inner classes. |
jgFages
previously approved these changes
Nov 15, 2022
…Selector when variables are added to constraint (like PropSat) during solving
mergify
bot
dismissed
jgFages’s stale review
Pull request has been modified.
|
/bench |
2 similar comments
|
/bench |
|
/bench |
|
/bench |
|
This PR is ready to be integrated but I'm waiting for #993 to be fully merged to rebase this branch and provide some information related to the performance. |
Step 1: `AbstractVariable`
…Constraints for the moment)
…Selector when variables are added to constraint (like PropSat) during solving
…Selector when variables are added to constraint (like PropSat) during solving
# Conflicts: # parsers/src/test/resources/xcsp/instances.csv # solver/src/main/java/org/chocosolver/solver/search/strategy/selectors/variables/AbstractCriterionBasedVariableSelector.java # solver/src/main/java/org/chocosolver/solver/variables/impl/AbstractVariable.java
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.
It appears that PR #889 has introduced a memory consumption issue.
Actually, the fact that propagators are not stored in an array (in
AbstractVariable) but in an array of arrays requires too much memory.Here are the memory consumption (in GB), estimated with
Runtimeand validated with JProfiler, of dummy models only composed of a very big number (1000000) of variables but no constraint:Where
4.10.11*denotes the proposal from this PR.The targeted class is
AbstractVariable, and the idea is mainly to create structures only when required.The current way of storing a variable's propagators depends on the events they react on.
In many cases, all events are not really covered so there is no need to create everything all the time.
I also had a look to other array creation (in solver/, ignoring /constraints/ directory first).