| Newsgroups |
gmane.ietf.rohc |
| Message-ID |
<[email protected]> |
Hello Lars-Erik
The main differences are that the proposed solution, TuCP does not
compress outermost IP header (IPout) and it focuses on packet
reordering problem as there can be many routers between the compressor
and decompressor.
The proposed draft defines header compression profiles for general
tunneling mechanisms. For example, in case of L2TP tunnel, ROHC does
not define a profile for compressing, IP/UDP/L2TP/PPP headers.
Moreover, in this example, IP is not to be compressed as it is
required for routing the packet to the final destination.
If we consider both the inner and outer encapsulation which is
IPout/UDP/L2TP/PPP/IPin/UDP/RTP,
currently, there is no profile defined in ROHC to compress this entire
encapsulation. The ROHC RTP profile can be used to compress part of
the encapsulation which is IPin/UDP/RTP. However, there is a problem
of packet reordering in IP tunnels which needs to be addressed when
ROHC (RFC 3095) is used over IP tunnels.
The proposed mechanism, TuCP, manages out of order packets, and thus
it enables the use of ROHC for inner IP packet headers efficiently.
Hope that answers your queries.
Regards
Priyanka Rawat
IT/TELECOM Bretagne
Lars-Erik Jonsson <[email protected]> a écrit :
> This submission does not make much sense to me, as ROHC already
> provides mechanisms for compression of tunneling headers (outer
> headers). To support tunneling was a requirement stated already by
> RFC 3096, and thus addressed by RFC 3095. Since then, compression of
> IP tunneling headers has been provided by all ROHC profiles.
>
> Can you explain what you mean ROHC is missing, not doing correcty or
> efficiently, or by which other means you think ROHC is not
> sufficient in terms of IP tunnel compression?
>
> BR
> /L-E
>
>
> ----- Original Message ----- From: <[email protected]>
> To: <[email protected]>
> Cc: <[email protected]>
> Sent: Monday, April 06, 2009 1:25 AM
> Subject: [rohc] Rohc over IP tunnel
>
>
>> Hello All
>>
>> We have submitted a new draft to IETF. This document proposes a
>> method to use ROHC over IP tunnel.
>>
>> The draft is available at
>> http://tools.ietf.org/search/draft-rawat-hc-tunneling-00
>>
>> Since we would like to continue this work further, it will be
>> interesting for us to receive feedback on the draft and initiate
>> discussion at IETF.
>>
>> Please let us know your opinion.
>>
>> Regards
>> Priyanka Rawat
>>
>> IT/TELECOM Bretagne
>>
>> _______________________________________________
>> Rohc mailing list
>> [email protected]
>> https://www.ietf.org/mailman/listinfo/rohc
>