The MPLS WG Archive

Cell Relay Retreat>MPLS WG Archive>month:2000-Jul> msg00293



[Date Prev][Date Next][Thread Prev][Thread Next]  
  [Date Index][Thread Index][Author Index][Subject Index]

IP header compression (was "process")

  • From: "Brijesh Kumar" <bkumar@ennovatenetworks.com>
  • Date: Thu, 20 Jul 2000 13:54:30 -0400
  • Cc: "'Eric Gray'" <EGray@zaffire.com>, "'George Swallow (E-mail)'" <swallow@cisco.com>, "'Vijay Srinivasan'" <Vijay.Srinivasan@cosinecom.com>, "'MPLS Mailing List (E-mail)'" <mpls@UU.NET>
  • Importance: Normal

Hi,

I think this discussion is going in the wrong direction.

IP header compression has definitely proved very useful, and is still
used in most wireless data devices. For instance CDPD devices use IP
header compression, and so do many other wireless (cellular) data
systems. The case for using mpls header compression is not that
straight forward due to this being a router to router mechanism,
though I do know instances where routers were placed on wireless
links. In a particular network in China, I used a VSAT 64 kbps common
channel to route IP traffic from about 60 cities. Each site had an IP
router, a few base stations and a switch. We  didn't use any dynamic
routing, and we only used static routing/bridging to simplify things,
and reduce routing protocol overhead. I am not sure if there are any
other situations which require a router to be placed on a slow
wireless link.

There are two issues: 1) understand technical mechanisms to do MPLS
header compression and 2) to justify the need of any one ever using
it. I think one can discuss the mechanisms (1) but the final decision
to proceed should only be taken after a case can be built for its real
use. If you put (2 i.e., utility) ahead of (1 i.e., mechanisms), I
think many other IDs in the IETF Internet Draft directory will also
fail to meet this criteria. I personally think we can briefly discuss
mechanisms for the time being, and freeze things after that unless the
authors of the draft can make a convincing case for its usage. Let us
have no-confrontation mid-path solution.

Cheers,

--brijesh
Ennovate Networks Inc.


> -----Original Message-----
> From: owner-mpls@UU.NET [mailto:owner-mpls@UU.NET]On Behalf Of Lou
> Berger
> Sent: Thursday, July 20, 2000 1:12 PM
> To: Eric Gray
> Cc: 'Lou Berger'; Eric Gray; George Swallow (E-mail); 'Vijay
> Srinivasan'; MPLS Mailing List (E-mail)
> Subject: RE: IP header compression (was "process")
>
>
>
> At 10:05 AM 7/20/00 -0700, Eric Gray wrote:
> >Lou,
> >
> >         I did, as you well know - since you were
> >copied. :-)
>
> Eric,
>          Your message was directed to me (on both subject
> line and in body)
> and to Eric R. on subject line.  Not the chairs.  The chairs
> were on cc'ed.