> On Wed, 4 Mar 1998, Marty Leisner wrote:
>
> > Any hard numbers on cpu time versus disk wait? Also how much
> > compression will be achieved...
>
> Disk transfer time is absolutely minimal, especially when
> compared to disk seek time...
> And disk seek time isn't at all affected by swap compression.
> If you want swap performance, clustering and read-ahead are
> the way to go.
it is affecting seek times .. a 2-times compression thing means the size
of the swap area is halved as well, thus average seeks are halved as well.
(if you use a separate swap device and swap fragmentation isnt too high,
ie. the swap set is a ~1.0 dimensional fractal ;)
-- mingo
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu