Re: Updated bootchart (rev1543)
Stéphane Graber <stgraber-GeWIH/[email protected]> Wed, 20 Jan 2010 10:31:08 -0500
| Newsgroups | gmane.linux.terminal-server.devel |
|---|---|
| Message-ID | <[email protected]> |
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Alkis Georgopoulos wrote:
> Since the nbd image is read-only, can't we tell the kernel to cache
> every block that it reads from it, so that it only needs to read each
> block once, if the RAM suffices?
It's already doing caching, actually, I find 20s for a full boot
perfectly acceptable especially as I think we'll easily get that down to
15s or even faster depending on the work Ubuntu's doing.
Just for the record, currently on the same hardware with a SSD, it takes
19s to boot, so LTSP is perfectly on track there.
>
> If the kernel already does that automatically, maybe *somehow* enabling
> readahead would help? Or maybe a bigger block size?
Well, that won't really work, I did some tests with ureadahead and it's
quite linked with the hardware + you need to boot a thin client to get
the binary file to include on your chroot, so that can't be scripted as
part of ltsp-build-client.
>
>
> Btw, I'd be interested to see how nbd-proxy affects low-end client
> performance, e.g. PIII / 64 RAM. Stéphane, when the optimizations are
> done, could you please notify me to do some testing?
Sure, I'll poke you. nbd-proxy shouldn't add much overhead other than
having data moving a lot between kernel and userspace. Currently
something must be wrong in the code as it's not killing the CPU and
gives us poor performance, must be a simple bug somewhere.
>
> -Alkis
- --
Stéphane Graber
Ubuntu developer
http://www.ubuntu.com
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iEYEARECAAYFAktXIbwACgkQjxyfqkjBhuwFrACfRuUhzawDBWqlsFYjQt1RvzmA
R+EAn1qT3VAKYG6at3aLOXjOW0coJ37G
=FBrY
-----END PGP SIGNATURE-----
------------------------------------------------------------------------------
Throughout its 18-year history, RSA Conference consistently attracts the
world's best and brightest in the field, creating opportunities for Conference
attendees to learn about information security's most important issues through
interactions with peers, luminaries and emerging and established companies.
http://p.sf.net/sfu/rsaconf-dev2dev
_____________________________________________________________________
Ltsp-developer mailing list. To un-subscribe, or change prefs, goto:
https://lists.sourceforge.net/lists/listinfo/ltsp-developer
For additional LTSP help, try #ltsp channel on irc.freenode.net