On Tue, 21 Mar 2000, Horst von Brand wrote:
> Where it counts, it will have been implemented. In any case, it will rarely
> have any huge impact. Most programs don't use sophisticated data structures,
> even less are able to cleanly distinguish between essential and
> non-essential (doing so is _much_ extra programmer effort, for no gain when
> the resources are available, and little if they aren't (it will mostly mean
> memory will definitely run out somewhat later)).
Some may be able to reduce memory usage by reducing the number of running
processes (apache), or the size of memory caches (squid, netscape).
No programming magic required. I do agree it's quite useless.
It helps only if you reach OOS for a very transient cause. A runaway
process will thankfully eat up every newly freed space in a short time.
OOM / OOS should be avoided. Keep your critical system under control.
You have good reasons to do that anyway. Running a critical server
unattended just because 'it does not overcommit' is nonsense.
> --
> Dr. Horst H. von Brand mailto:vonbrand@inf.utfsm.cl
> Departamento de Informatica Fono: +56 32 654431
> Universidad Tecnica Federico Santa Maria +56 32 654239
> Casilla 110-V, Valparaiso, Chile Fax: +56 32 797513
.TM.
--
____/ ____/ /
/ / / Marco Colombo
___/ ___ / / Technical Manager
/ / / ESI s.r.l.
_____/ _____/ _/ Colombo@ESI.it
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Thu Mar 23 2000 - 21:00:33 EST