[e2e] Can we revive T/TCP ?
touch at ISI.EDU
Tue Apr 4 15:24:00 PDT 2006
-----BEGIN PGP SIGNED MESSAGE-----
Bob Briscoe wrote:
> There's a wrinkle that argues against demux at the TCP layer, and
> instead argues for demux at the app layer in these cases: Equal Cost
> Multipath Routing (ECMP).
> Some vendor implementations of ECMP forward flows deterministically on
> the same path by hashing packets based on ip addr & port numbers to
> choose which interface to forward out of. Others ignore port numbers.
Whether that matches your intent is determined by your definition of "flow".
> But changing port numbers can mean your packet ends up crossing
> different routes. So, changing port numbers can imply the congestion
> information learned from the previous connection is no longer valid.
There are many forwarding algorithms that use information beyond that of
only endpoint address, generally called 'policy routing'. Yes, they
break the notion of transport flows - for almost any definition of
transport flow anyway.
> The alternative is to be more prescriptive by saying ECMP algorithms
> MUST NOT use port numbers, but these algs are already out there.
Even if port numbers aren't used, source address can be, so can other
fields (IPv6 flowspec, etc.).
> Another answer is encryption, to stop these boxes fishing around where
> they have no business anyway.
You can encrypt anything you don't want the network to see - or possibly
forward differentially on. All that does is encourage differential
forwarding on SPIs, though.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
-----END PGP SIGNATURE-----
More information about the end2end-interest