Re: [RFC][PATCH] O(1) Entitlement Based Scheduler

[Peter Chubb]

>>>>> "Paul" == Paul Jackson <pj@xxxxxxx> writes:


Paul> Is there anyway to provide a mechanism that would support
Paul> administering a system as follows:

Paul>   1) Users get so much CPU usage allowed, determined by an upper
Paul> limit on a running average of the combined CPU usage of all
Paul> their tasks, with a half life perhaps on the order of minutes.

Paul>   2) They can nice their tasks up and down, within a decent
Paul> range, as they will.

Paul>   3) But if they push too close to their allowed limit, all
Paul> their tasks get reined in.  The relative priorities within their
Paul> own tasks are not changed, but the priority of their tasks
Paul> relative to other users is weakened.

This is exactly what the commercial product ARMtech does.  The EBS
that Aurema have just released as open source is (a small) part of the
commercial product.

See http://www.aurema.com

Peter C (an ex-employee of Aurema)

--
Dr Peter Chubb  http://www.gelato.unsw.edu.au  peterc AT gelato.unsw.edu.au
The technical we do immediately,  the political takes *forever*
-
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/