On Wednesday 20 December 2006 7:14 am, Kristian Kielhofner wrote:

<snip good stuff>

>         As far as traffic shaping goes in general...  Yes, to do effective
> traffic shaping with VoIP services on "standard" internet connections,
> you really should limit the overall transfer capacity to about %90 of
> what is capable on the link.
>
>         Just like Darren described, if you don't artificially limit the
> speed of the link your ISP will, and their send queues are not tuned like
> yours are...  So, if you run at %100, and they start queuing, they are
> going to use whatever algorithm that they have (probably a simple FIFO with
> a buffer) to schedule outgoing packets.  This isn't what you want! Limiting
> the speed of the link prevents this queuing and will give you a better
> experience for VoIP (and any other service that uses large amounts of
> relatively small packets - gaming for instance).

I second the notion. I haven't used any traffic shaping on my little boards, 
but I've used Wonder Shaper a lot on old PC firewalls for DSL connections. 
Even though you're throttling down to 90%, you'll experience faster downloads 
and better performance because of less queueing. It seems paradoxical, but 
the queueing is what hurts, so minimizing that makes a noticeable difference.

Thanks Kristian for the explanation of kernel timing, I think I'm finally 
starting to get it. That's something that always seems to be lurking in wait, 
just waiting for a worst moment to leap out and make a mess of something.

-- 
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Carla Schroder
Linux geek and random computer tamer
check out my Linux Cookbook! 
http://www.oreilly.com/catalog/linuxckbk/
best book for sysadmins and power users
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

_______________________________________________
Astlinux-users mailing list
Astlinux-users@lists.kriscompanies.com
http://lists.kriscompanies.com/mailman/listinfo/astlinux-users

Donations to support AstLinux are graciously accepted via PayPal to [EMAIL 
PROTECTED]

Reply via email to