Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!seismo!caip!nike!ucbcad!ucbvax!KA9Q.BELLCORE.COM!karn
From: karn@KA9Q.BELLCORE.COM (Phil Karn)
Newsgroups: mod.protocols.tcp-ip
Subject: Re: TCP retransmission efficiency
Message-ID: <8607280403.AA00258@ka9q.bellcore.com>
Date: Mon, 28-Jul-86 00:03:39 EDT
Article-I.D.: ka9q.8607280403.AA00258
Posted: Mon Jul 28 00:03:39 1986
Date-Received: Mon, 28-Jul-86 04:41:56 EDT
Sender: daemon@ucbvax.BERKELEY.EDU
Organization: The ARPA Internet
Lines: 12
Approved: tcp-ip@sri-nic.arpa

Forget the Nagle algorithm for the moment; I confused the issue by
bringing it up.

Assuming ACKs aren't lost, unnecessary retransmissions are avoided
as long as you have "first only" retransmission at the sender plus
"in window" acceptance at the receiver.

As I said, "ack retransmission" (i.e., acking acks) is a last-ditch
approach for use over very unreliable links. I'm not sure it'd be
at all appropriate at the transport (host-host) layer.

Phil