----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/8108/#review13626 -----------------------------------------------------------
src/slave/cgroups_isolation_module.hpp <https://reviews.apache.org/r/8108/#comment29212> doc please src/slave/cgroups_isolation_module.cpp <https://reviews.apache.org/r/8108/#comment29208> Can you doc the algorithm? I think the idea is to always have only 1 partial cpu globally? src/slave/cgroups_isolation_module.cpp <https://reviews.apache.org/r/8108/#comment29210> The order of cpus is guaranteed here because usage always contains all the cpus in the system? src/slave/cgroups_isolation_module.cpp <https://reviews.apache.org/r/8108/#comment29211> Actually thinking a bit about it, fragmentation means sometimes you cannot do an allocation? Failing here seems drastic. We should probably find a way to reject the grow without failing. src/slave/cgroups_isolation_module.cpp <https://reviews.apache.org/r/8108/#comment29207> 'delta' in a log line probably doesn't make sense to users. how about just say 'by' ? src/slave/cgroups_isolation_module.cpp <https://reviews.apache.org/r/8108/#comment29206> ditto src/slave/cgroups_isolation_module.cpp <https://reviews.apache.org/r/8108/#comment29213> nice src/slave/cgroups_isolation_module.cpp <https://reviews.apache.org/r/8108/#comment29214> why don't you pull info->cpuset.get() out into a variable. Looks like its used a bunch here. some of the statements below might actually fit into one line then. - Vinod Kone On Nov. 19, 2012, 11:03 p.m., Ben Mahler wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/8108/ > ----------------------------------------------------------- > > (Updated Nov. 19, 2012, 11:03 p.m.) > > > Review request for mesos, Benjamin Hindman and Vinod Kone. > > > Description > ------- > > This is the first pass at adding cpuset isolation for pinning cgroups to cpus. > > We decided to start with a simplistic grow/shrink allocation technique, as > such this initial technique: > -Does not take cache locality into account. > -Does not actively fight fragmentation*, but does a good job at preventing > it in many cases, given it's simplicity. > -Note that when cpus resource requests are integral (non-fractional), then > fragmentation does not occur. > > *By fragmentation, I'm referring to the case where we've spread a cgroup over > more cpus than necessary, due to other cgroups sharing the same cpus. > High fragmentation would mean a lot of shared cpus across cgroups. > No fragmentation would mean each cgroup has a unique set of cpus. > > I've punted on documenting the pitfalls of this technique, wiring up the > handler, and adding tests for now. > > Note that this is diffed off of benh's changes: > https://reviews.apache.org/r/8058/ > https://reviews.apache.org/r/8059/ > > > Diffs > ----- > > src/linux/proc.hpp 27e15bf8695aa694b0d5bdb6881b9fa55a447528 > src/slave/cgroups_isolation_module.hpp > 9f80fc5a969b959b34eaea4cac40700662d7f8b2 > src/slave/cgroups_isolation_module.cpp > 8211618d7729350654e2d17946c5b912ed9dda6a > third_party/libprocess/include/stout/stringify.hpp > dcc5b95a54e6f34f93867e015d8c855fd7d6f950 > third_party/libprocess/include/stout/strings.hpp > 914c280a994733764957d19f37b48d151bb93778 > > Diff: https://reviews.apache.org/r/8108/diff/ > > > Testing > ------- > > None as of yet. > > > Thanks, > > Ben Mahler > >
