Re: [PATCH] TCP Vegas implementation available

Aaron Tiensivu (mojomofo@ctechnix.com)
Tue, 20 Jul 1999 21:00:01 -0400


> results indicate Vegas can help for low-bandwidth-delay paths like DSL
> where the sender is constantly over-running buffers, or
> high-bandwidth-delay WAN paths where the receiver isn't sending SACKs and
> the sender is wasting a lot of time recovering from losses.

So far so good on my Multia which acts as the internet server/masqbox with
Vegas alpha/beta set at 3, on a 53k analog modem connection. I've noticed
much "smoother" connections. Don't have any tcpdumps to prove or disprove it
but that's just my 'ball-park' feeling of it.

This is under 2.2.10-ac12 and the patch applies cleanly (with a few offset
lines but that's no biggie)

I sent out a mass-email to about 20 people and it floored the modem lights with
SMTP/Auth connections. That was the biggest improvement (which would make sense
) I've noticed so far.

> I'd love to get any feedback people have on the code or performance
> results people see from Vegas.

I don't have hard numbers at the moment but it looks golden here.

--
Which is worse, ignorance or apathy? Who knows? Who cares?
I'm not a perfectionist, but I'm working on 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/