----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/37913/#review97346 -----------------------------------------------------------
Ship it! Looks great, thanks! I'll will get this commmitted for you shortly, left some comments for small adjustments I'll make before committing. src/master/allocator/mesos/hierarchical.hpp (line 216) <https://reviews.apache.org/r/37913/#comment153219> You only used this in one location below? In general we've avoided typedefs since they require a bit more non-local reasoning at the point in which they are being used (i.e. what is a FilterSet? Turns out it's a hashset of Filter pointers). src/master/allocator/mesos/hierarchical.hpp (line 1084) <https://reviews.apache.org/r/37913/#comment153261> Can we keep this newline? - Ben Mahler On Aug. 28, 2015, 11:46 p.m., James Peach wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/37913/ > ----------------------------------------------------------- > > (Updated Aug. 28, 2015, 11:46 p.m.) > > > Review request for mesos and Ben Mahler. > > > Bugs: MESOS-3052 > https://issues.apache.org/jira/browse/MESOS-3052 > > > Repository: mesos > > > Description > ------- > > When frameworks refuse a lot of resources, the list of filters gets > long. Since the filters are per-slave, > HierarchicalAllocatorProcess::isFiltered spends a lot of time just > comparing SlaveID (which tend to be long strings). Eliminate this > whole problem by organizing the filters by SlaveID in the first > place. > > > Diffs > ----- > > src/master/allocator/mesos/hierarchical.hpp > 38f8fd2c84314bb3731684d0e9795cb4f50a227e > > Diff: https://reviews.apache.org/r/37913/diff/ > > > Testing > ------- > > make check on CentOS 6 w/ devtoolset-3. > > > Thanks, > > James Peach > >