A profiling branch has landed in Loggerhead trunk, which we've been using to profile Loggerhead. I have a branch that uses a memory middleware to monitor memory usage. We believe this issue has to do with how we generate a log for a file. If we can use bzr's way of doing it, I think we can benefit from their optimizations.
A profiling branch has landed in Loggerhead trunk, which we've been using to profile Loggerhead. I have a branch that uses a memory middleware to monitor memory usage. We believe this issue has to do with how we generate a log for a file. If we can use bzr's way of doing it, I think we can benefit from their optimizations.