Re: [aqm] Question re draft-baker-aqm-recommendations recomendation #2

On Apr 29, 2013, at 7:46 AM, Linda Dunbar <linda.dunbar@huawei.com>
 wrote:

> Generally agree with signaling to an endpoint using both dropping and ECN, except ECN marking may not be possible in MPLS networks because MPLS ECN uses EXP bits that are used to mark Priority in many deployed networks. 

Well, "impossible" and "requires some effort" are two different things. I suspect that the LSP endpoint, and any MPLS endpoint that adds or drops labels in the stack, could copy MPLS-ECN from the outer header to the inner header as it does so, as is recommended for IP/IP tunneling. It does have a syntactic issue, in that MPLS has relatively few bits to work with. But I think the semantics are clear enough.

> I think that it is important to emphasize that ECN marking for end points to notify source nodes don't work well in very large networks. 
> 
> Since today's network is tunnels over tunnels (multiple layers of tunnels, e.g. MPLS over PW),  the tunnel scoped ECN marking can work well if the outmost tunnel doesn't go over too many hops and subsequent tunnels control the ingress rate. 

Thanks

> Linda Dunbar
> 
>> -----Original Message-----
>> From: aqm-bounces@ietf.org [mailto:aqm-bounces@ietf.org] On Behalf Of
>> Fred Baker (fred)
>> Sent: Monday, April 29, 2013 9:00 AM
>> To: aqm@ietf.org
>> Subject: [aqm] Question re draft-baker-aqm-recommendations
>> recomendation #2
>> 
>> Do we generally agree with the recommendation of
>> http://tools.ietf.org/html/draft-baker-aqm-recommendation-01#section-
>> 4.2? This is the question of signaling to an endpoint using both
>> dropping and ECN.
>> 
>> 
>> _______________________________________________
>> aqm mailing list
>> aqm@ietf.org
>> https://www.ietf.org/mailman/listinfo/aqm