Re: [p2pi] Charter and problem statement

[Sebastian Kiesel <sebastian.kiesel at nw.neclab.eu>]

On Fri, Jul 11, 2008 at 11:10:41AM -0400, Laird Popkin wrote:
> For this level of discussion, I agree that caches should simply be
> viewed as a peer in ALTO.

What happens if we consider caches and rare content, i.e., content where
the likelyhood is high, that there is a cache miss when we ask for it
and that the cache will never deliver a second copy of it? It might not
be wise to select a cache which is in the "opposite direction" of the
original content, even if connectivity to the cache is "better"
(according to whatever criteria). If the cache tries to fetch the
content the request may pass by the client and go through the same "bad"
link plus twice through the "good" one.

Cache ====Link1(good)==== Client -----------Link2(bad)----------- Server


This becomes even more important if we want to generalize ALTO and use
it, e.g., to select media relays for VoIP NAT traversal:

                       UserA                     UserB
                         |                         |
                        NAT                       NAT
                         |                         |
Relay1 ====L1(medium)====+---------L2(bad)---------+==L3(good)== Relay2
                         |
                        NAT
                         |
                       UserC

UserA should tell the ALTO service not only that it needs "a VoIP relay"
but also that he wants to talk to UserB.



Thanks,
Sebastian


-- 
Sebastian Kiesel            mailto:sebastian.kiesel at nw.neclab.eu
Network Research Division   tel:+49-6221-4342-232   fax:+49-6221-4342-155
NEC Laboratories Europe     Kurfuerstenanlage 36, 69115 Heidelberg, Germany
--
NEC Europe Limited          Registered in England 2832014
Registered Office           NEC House, 1 Victoria Road, London W3 6BL
_______________________________________________
p2pi mailing list
p2pi at ietf.org
https://www.ietf.org/mailman/listinfo/p2pi