[ltp] swsusp vs. Hibernate?

[Bret Comstock Waldow]

--nextPart1288811.hv3U3bokgT
Content-Type: text/plain;
  charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline

On Fri, 11 Nov 2005 02:58, Aaron Mulder wrote:
> On 11/9/05, Bret Comstock Waldow <bwaldow@woosh.co.nz> wrote:
> > So, why not just dedicate a larger swap partition (existing swap + 1G),
> > which you'll get more use out of than just for hibernating?
>
> This is totally subjective, but it doesn't seem like the normal kernel
> implementation attempts to shrink only until it fits in the available
> swap space.  It seems to try to shrink as much as it possibly can and
> then goes and applies whatever it's left with to swap space.  So
> adding more swap space doesn't seem likely to shorten the "shrinking"
> cycle.  Am I wrong?

I will admit that I don't know about that.

I am pretty sure however, that neither a separate swap partition, nor Redis=
afe=20
will speed this up.  Perhaps swsusp2 might - if the compression+writing is=
=20
quicker than the writing alone.

Bret

--nextPart1288811.hv3U3bokgT
Content-Type: application/pgp-signature

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.1 (GNU/Linux)

iD8DBQBDc4N6922F8D4woqURAmM6AJ98z/YOQa5WNriGikYpMTLLYIn+awCcCvZp
+iy5PoPumlH0Bcff8QHvWYU=
=WCt7
-----END PGP SIGNATURE-----

--nextPart1288811.hv3U3bokgT--