Matthew Jacobs has posted comments on this change. Change subject: IMPALA-3291. Pause monitor thread should check tcmalloc performance ......................................................................
Patch Set 1: (3 comments) http://gerrit.cloudera.org:8080/#/c/2707/1/be/src/common/init.cc File be/src/common/init.cc: Line 155: gotten blocked just "blocked" Line 160: 128 * 1024 Is this number meaningful at all (e.g. for tcmalloc)? Can you touch on this in the comment? Line 167: MarkThreadIdle > Todd, thanks for answering my question in person (pointing out that this is Does that mean the next time around the thread cache will be created when we call malloc? Maybe that's fine, but I wonder if we might be hitting other "non organic"/unrealistic delays as a result of creating/destroying thread caches? My guess is it's OK but just curious what your thoughts are. -- To view, visit http://gerrit.cloudera.org:8080/2707 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-MessageType: comment Gerrit-Change-Id: I40516456dda9ea6cfce3f2074ab9a4b25087c4bd Gerrit-PatchSet: 1 Gerrit-Project: Impala Gerrit-Branch: cdh5-trunk Gerrit-Owner: Todd Lipcon <[email protected]> Gerrit-Reviewer: Bharath Vissapragada <[email protected]> Gerrit-Reviewer: Dan Hecht <[email protected]> Gerrit-Reviewer: Matthew Jacobs <[email protected]> Gerrit-HasComments: Yes
