[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: Executive summary of: RE: hard questions: request routing



Title: RE: Executive summary of: RE: hard questions: request routing


> -----Original Message-----
> From: Cain, Brad [mailto:bcain@cereva.com]

> > 3. Abbie's proposal
> actually, isn't this similar in concept to #2

No I do not think so, in particular with respect to your quote

> ... the only difference here is
> that the final circuit (e.g. the set of hops) is returned
> to the originator so that the final hop can be returned
> (and of course intermediate state isn't kept)


> > 4. Abbie's proposal as I understood it first..... .

> i assume that the authoritative CDN makes all decisions (based
> on the link state information)?  #1 would be the degenerate case?
>

yes and no. You can restrict it to the authoritative CDN or allow for rerouting


> should the authoritative request routing CDN always make the
> final decision or should it be allowed to "hand-off" the decision? 

I think it should be able to hand-off the decision.

> if the decision is centralized then it makes the problem fairly easy...
> you could make a strong argument for
> centralized decisions due to the business relationships in
> content distribution.. furthermore you can make an argument
> that routing the request introduces too much delay and that
> the authoritative domain should make the decision
>

The ability to route should be there.


> i propose a 5th solution which is:
>
>       - topology advertisements are distributed/flooded in a loop
>       free manner
>       - routing decisions are NOT SPECIFIED.  in the common case the
>       authoritative domain makes the final decision with all of the
>       topology information ( = easy loop freedom!)

-- above are similar to the matrix appraoch


>       - requests that are routed to another CDN must not be
>       routed further (ie. restricted two-level hierarchy)

-- this is option 1


abbie