[e2e] Is a non-TCP solution dead?
George Michaelson
ggm at apnic.net
Wed Apr 2 15:41:32 PST 2003
On Wed, 02 Apr 2003 22:44:20 +0100 Jon Crowcroft <Jon.Crowcroft at cl.cam.ac.uk>
wrote:
> where's the code
> repeat after me - if its a delta, and the performance gain is more
> than epslion, and the pain is less than iota
> why doesnt everyone use it>?
>
> tcp = proof of the pudding.
> cheers
>
> jon
ftam was shown to get 75% of ftp throughput. that was one implementation.
given time, I suspect it could have achieved the same throughput. As to
whether the transport or its end-to-endness could have scaled better to
current network behaviours I am less sure, but I think there stands the
example that even approaching the slope of the line, if the pain is not less
than iota, its just not going to happen.
interesting to look at what pragmatic decisions we make in apps-space.
Hands up how many of you use rsync and rsync/ssh to do things
which http and ftp claim to 'own' -?
applications-space datagram protocol with checksum over end-to-end reliable
transport wins again!
-George
--
George Michaelson | APNIC
Email: ggm at apnic.net | PO Box 2131 Milton QLD 4064
Phone: +61 7 3367 0490 | Australia
Fax: +61 7 3367 0482 | http://www.apnic.net
More information about the end2end-interest
mailing list