Sailesh Mukil has posted comments on this change. Change subject: PREVIEW: IMPALA-2550 Introduce query-wide execution context. ......................................................................
Patch Set 3: (1 comment) http://gerrit.cloudera.org:8080/#/c/3686/2/be/src/service/fragment-mgr.h File be/src/service/fragment-mgr.h: PS2, Line 80: e corresponding > It sounds like you're describing basically shared_ptr acquire / release sem A question. What happens if one fragment arrives at this backend, finishes executing and therefore destroys the state and leaves. Now a second fragment arrives late due to some reason. Does the second fragment end up setting up the QueryExecState all over again? Or is this scenario of a fragment arriving much later like described above not realistic? I thought putting forward this point could help in the decision of shared_ptr vs ref counts. -- To view, visit http://gerrit.cloudera.org:8080/3686 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-MessageType: comment Gerrit-Change-Id: I9c513d08f718699ba0c4cdb90c117aaecf95d7fc Gerrit-PatchSet: 3 Gerrit-Project: Impala Gerrit-Branch: cdh5-trunk Gerrit-Owner: Lars Volker <[email protected]> Gerrit-Reviewer: Alex Behm <[email protected]> Gerrit-Reviewer: Dan Hecht <[email protected]> Gerrit-Reviewer: Henry Robinson <[email protected]> Gerrit-Reviewer: Lars Volker <[email protected]> Gerrit-Reviewer: Marcel Kornacker <[email protected]> Gerrit-Reviewer: Sailesh Mukil <[email protected]> Gerrit-Reviewer: Tim Armstrong <[email protected]> Gerrit-HasComments: Yes
