[ https://issues.apache.org/jira/browse/TINKERPOP-1822?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16541080#comment-16541080 ]
ASF GitHub Bot commented on TINKERPOP-1822: ------------------------------------------- Github user krlohnes commented on a diff in the pull request: https://github.com/apache/tinkerpop/pull/838#discussion_r201900734 --- Diff: gremlin-core/src/main/java/org/apache/tinkerpop/gremlin/process/traversal/step/branch/RepeatStep.java --- @@ -273,11 +300,40 @@ public RepeatEndStep(final Traversal.Admin traversal) { super(traversal); } + final LinkedList<Traverser.Admin<S>> stashedStarts = new LinkedList<>(); --- End diff -- @spmallette Those results are definitely...interesting to say the least. I think the tests themselves are reasonable, though, as a comment, I'm not typically using a repeat that's going to be able to utilize the `RepeatUnrollStrategy`. At least not for what I originally started investigating this for. That said, I took a step back and worked on performance between the BFS and DFS, and have gotten them much closer. On my local machine that BFS test was returning 889 from the counter. With the latest commit I added, DFS is returning 758. That's obviously not coming close to the default "let the strategies do their thing" performance, but it's significantly better than the ops counts being in the teens for DFS. Given that I was expecting slightly degraded performance with DFS, I think this is in a much better place performance wise. > Repeat should depth first search > -------------------------------- > > Key: TINKERPOP-1822 > URL: https://issues.apache.org/jira/browse/TINKERPOP-1822 > Project: TinkerPop > Issue Type: Improvement > Components: process > Affects Versions: 3.3.0, 3.2.6 > Reporter: Robert Dale > Priority: Major > Attachments: bfs.jfr, default.jfr, dfs.jfr > > > Perhaps optionally. > See also: > * https://groups.google.com/forum/#!topic/gremlin-users/gLSLxH_K-wE > * https://github.com/apache/tinkerpop/pull/715 -- This message was sent by Atlassian JIRA (v7.6.3#76005)