[
https://issues.apache.org/jira/browse/TINKERPOP-2919?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17707423#comment-17707423
]
ASF GitHub Bot commented on TINKERPOP-2919:
-------------------------------------------
vkagamlyk commented on PR #2009:
URL: https://github.com/apache/tinkerpop/pull/2009#issuecomment-1492596313
As far as I understand everything works as before, but without multiple
recalculations.
The only improvement I see is caching of `stepRanking`. It used in `apply`,
but probably `getStepRank` and `getMaxStepRank` also can benefit from caching.
Another possible approach to add 'getRank()' to the base Step interface and
remove related code from `FilterRankingStrategy`.
LGTM, VOTE +1
> Improve performance of FilterRankingStrategy for deeply nested traversals
> -------------------------------------------------------------------------
>
> Key: TINKERPOP-2919
> URL: https://issues.apache.org/jira/browse/TINKERPOP-2919
> Project: TinkerPop
> Issue Type: Improvement
> Components: process
> Affects Versions: 3.5.5
> Reporter: Stephen Mallette
> Priority: Major
>
> There is extremely heavy recursion in {{FilterRankingStrategy}} so when it
> encounters traversals with a lot of depth it can take an extremely long time
> to execute.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
