Re: [alto] FW: I-D Action:draft-stiemerling-alto-deployments-05.txt

["Schmidt, Christian 1. (NSN - DE/Munich)" <[e-mail]>]

Hi Martin,

am I right, that mobile peers can be identified by the Alto Server?
If yes, perhaps the Alto Server can make something like a fast check of
the mobile peer position in the map - before offering him as source
peer.
The result of this check must not be perfect, the result should be valid
in a majority of cases.

Br
Christian








-----Original Message-----
From: ext Martin Stiemerling [mailto:[e-mail]] 
Sent: Friday, October 29, 2010 9:48 AM
To: Schmidt, Christian 1. (NSN - DE/Munich); [e-mail]
Subject: RE: [alto] FW: I-D
Action:draft-stiemerling-alto-deployments-05.txt

Hi Christian, 

You are raising a good point, which is not covered by the deployments
draft.

Mobile peers will also challenge the map-based approach, unless the
preference information always indicates to turn away from these types of
peers. This could be the most likely case, i.e., the ISP will configure
to rank those mobile peers really bad, so that other peers will not pick
them. 

The challenge is in the case where there is more specific configuration
needed. For instance, static mobile peers (e.g., UMTS stick in home
gateway), might get a slightly better rating, as well as, mobile peers
on HDSPA.

However, the rating for such a mobile peer may change rapidly, if the
peer starts moving and even changes the access technology (e.g. HSDPA to
EDGE). The IP address of the peer will remain the same but the rating
will dramatically change. 

I guess this is hard to manage with the map-based approaches. 

Thanks and Cheers to Munich

  Martin


> -----Original Message-----
> From: Schmidt, Christian 1. (NSN - DE/Munich)
> [mailto:[e-mail]]
> Sent: Friday, October 29, 2010 9:18 AM
> To: Martin Stiemerling; [e-mail]
> Subject: RE: [alto] FW: I-D Action:draft-stiemerling-alto-deployments-
> 05.txt
> 
> Hi Martin,
> 
> a short question concerning the map-based approach:
> 
> "The main assumption for map-based approaches is that the information
> provided in these maps is static for a longer period of time, where
> this
> period of time refers to days, but not hours or even minutes. This
> assumption is fine, as long as the network operator does not change
any
> parameter, e.g., routing within the network and to the upstream peers,
> IP address assignment stays stable (and thus the mapping to the
> partitions).  However, there are several cases where this assumption
is
> not valid, as:
>    1.  ISPs reallocate IPv4 subnets from time to time;
>    2.  ISPs reallocate IPv4 subnets on short notice;
>    3.  IP prefix blocks may be assigned to a single DSLAM which serves
> a
> variety of access networks."
> 
> What about mobile peers? How do they fit into a map-based approach?
> 
> Christian
> 
> 
> 
> 
> 
> 
> 
> 
> -----Original Message-----
> From: [e-mail] [mailto:[e-mail]] On Behalf
Of
> ext Martin Stiemerling
> Sent: Friday, October 29, 2010 8:37 AM
> To: [e-mail]
> Subject: [alto] FW: I-D Action:draft-stiemerling-alto-deployments-
> 05.txt
> 
> Dear all,
> 
> There is a slightly updated version of the deployment considerations
> draft.
> 
> Rich Alimi, Richard Yang, Xianghui Sun, Sebastian and me will come up
> with a joint proposal how to evolve the document at the next IETF.
> 
> Any comments are appreciated!
> 
>   Martin
> 


[e-mail]

NEC Laboratories Europe - Network Research Division
NEC Europe Limited | Registered Office: NEC House, 1 Victoria Road,
London W3 6BL | Registered in England 2832014