On 10/23/07, Srivatsa Vaddagiri <vatsa@xxxxxxxxxxxxxxxxxx> wrote: > > agreed, we need to be reporting idle time in (milli)seconds, although > the requirement we had was to report it back in percentage. I guess the > percentage figure can be derived from the raw idle time number. > > How about: > > idle time = t4-t3 (effectively sleep time) > > in the above example? > > > It doesn't seem quite right to me that a cgroup's idle time metric be > > affected by the activity of other cgroups on the machine, > > I don't see how the idle time metric defined above (t4-t3) can be > affected by other cgroup activity, unless the execution pattern of one > cgroup is dependent on the others. If the other cgroups are busier, and t1-t2 is longer, then the cgroup will get to the point where it's ready to sleep later in wallclock time, and t4-t3 will be shorter in absolute terms. If there were no other cgroups running, then presumably the sleep time would actually be the sum of those three periods. Even so, I guess you're right that t4-t3 is the most appropriate thing to report, as long as people realise that it's a bit of a fuzzy value. > I think primarily for systems management tools to report back various > statistics about a OpenVZ/VServer-like container (just like top reports stats > for a OS currently). Let me check if there are other uses envisoned for > it. Sorry, I didn't mean "how will you report it to users?", I meant "what kinds of useful information will the users be able to get from it?" Paul _______________________________________________ Containers mailing list Containers@xxxxxxxxxxxxxxxxxxxxxxxxxx https://lists.linux-foundation.org/mailman/listinfo/containers
