[MUD-Dev] [TECH] TCP fundamental throughput limits?

Jeremy Noetzelman jjn at kriln.com
Mon Oct 27 21:45:18 New Zealand Daylight Time 2003

On Fri, 24 Oct 2003, Miroslav Silovic wrote:

> On an unreliable connection, packet loss is measured in
> percentage, actually. This means that the TCP connection is in the
> state of permanent recovery. My own experience is that at 400 ms
> RTT (trans-atlantic satellite link), 15-20% packet loss drops the
> speed to well under 100 bytes/sec, making it completely unusable
> ragardless of the actual link capacity.

For the user, certainly.  From a stack perspective and a protocol
perspective, percentage loss is irrelevant.  You can always, as a
user or an application, measure packetloss in a TCP connection as a
percentage.  However, to the stack internals, which is what the
original poster was wondering about, percentages are irrelevant and
not calculated.

> In the particular simple recovery scheme you described, when
> packet loss is sufficiently high that the mean time between packet
> drops goes below decongestion avoidance window size, the transfer
> rate exponentially decreases to zero (because the slow restarts
> happen more often than speed doublings).

This is absolutely true, and one of the major problems with TCP
implementations over poor quality circuits.
MUD-Dev mailing list
MUD-Dev at kanga.nu

More information about the MUD-Dev mailing list