CGroup subsystem for enforcing hard cpu rate limits

[Chris Brox]

Back in 2006 Srivatsa Vaddagiri summarised proposals for CPU controllers,
which variously offered to limit or guarantee CPU for a task group. Some
of these supported both soft and hard limits.

Now in 2008 we have FAIR_CGROUP_SCHED in mainline, which supports soft CPU
limits per control group. Great stuff.

Unfortunately I am keenly looking for a hard CPU limit implementation -
i.e. one that would constrain a control group to use no more than x% of
the CPU, even when it is otherwise idle.

Has the hard limit idea been discarded, or is it still a planned feature
(of either the Fair scheduler or a new subsystem)?


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/