Re: my broken TCP is faster on broken networks [Re: Very poor TCP/SACK performance]
Savochkin Andrey Vladimirovich (saw@msu.ru)
Thu, 10 Sep 1998 09:42:32 +0400
On Thu, Sep 10, 1998 at 03:00:45AM +0200, Andrea Arcangeli wrote:
> On 9 Sep 1998, Mark Gray wrote:
>
> >OpenBSD uses a window of 16384 and I find that that gives me a very
> >good transfer rate with linux-2.0.* without ever causing stalls which
> >will occur off and on when using the default Window.
>
> I can reproduce the stall very very well in ppp. The problem is that some
> network from here to my Unversity are really congestioned, I think that
> because they drop many many TCP packets (or transmit it very out of
> order after a lot of seconds).
>
> The _only_ way to increase performance is to decrease the tp->rto. I tried
> force the kernel to retransmit the packet very fast (sure in less time
> than the rtt hehehe ;-) and my connection to the University now is
> perfectly responsive (cool ;-). I don' t stall anymore. Before my hack to
> tcp_reset_xmit_timer() I had to wait 10/15 sec before ssh asks me the
> password. Now I get the password prompt after some sec (and my RX modem
> line is always ON ;-). The interactive performance are perfect now with my
> brute hack.
You've forgotten one point.
If all people hacked their kernels as you did you aren't able to connect
to your University at all.
Your changes increase the performance of your connection by the price
of much more congestion for other connections.
The retransmit rules were designed for better total performance of
all connections. The retransmit timeout is set to a value
giving a very low probability of considering a traveling packet as lost.
As far as I remember the value of the timeout is set to
round trip time plus its mean deviation multiplied by 4.
What you've done is considered as network abuse in many places
and it's a good way to be disallowed to use the network.
Best regards
Andrey V.
Savochkin
-
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/faq.html