On Thu, Feb 02, 2012 at 02:55:05PM +0100, stefano previdi wrote:
> On Feb 2, 2012, at 1:05 PM, Sebastian Kiesel wrote:
> > On Thu, Feb 02, 2012 at 07:45:45AM +0100, stefano previdi wrote:
> >> On Feb 1, 2012, at 11:49 PM, Sebastian Kiesel wrote:
> >>>> Not that I'm advocating this but it all depends on where 
> >>>> do you want to place the complexity of topology hint: inside ALTO 
> >>>> server (ECS like of approach) or inside the application (Maps that 
> >>>> allows you to compute paths/trees).
> >>> 
> >>> Maybe part of the confusion is that we need a clear definition what
> >>> end-to-end is. I'd say that ALTO should always give the end-to-end
> >>> costs for sending IP packets between two endpoints using "normal"
> >>> IP-layer forwarding.  Based on that knowledge one can of course
> >>> build efficient paths/trees that involve proxies, media relays
> >>> or the like (i.e., this higher-layer path is a concatenation of
> >>> several IP-layer end-to-end paths). Does this make sense?
> >> 
> >> 
> >> why do you believe alto should only deliver end-to-end costs ?
> >> In a slightly different context it could even deliver a path.
> > 
> > what exactly do you mean by "deliver a path"?
> > 
> > And would that require that a host and/or the ALTO server can influence
> > how packets travel through the network (source routing, take actively
> > part in OSPF/BGP/... , etc)?
> 
> 
> well, I don't go that far. However, an application may ask for network 
> guidance and get some preferences (rankings) computed by an alto server 
> that goes along with some form of path description associated to that 
> preference.

and what would the application do with that additional information?

if it is used for re-calculating ALTO's preferences, why can't that
be done in the ALTO server in the first place?

Thanks,
Sebastian
_______________________________________________
alto mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/alto

Reply via email to