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

RE: hard questions: request routing



Title: RE: hard questions: request routing


> -----Original Message-----
> From: Oliver Spatscheck [mailto:spatsch@research.att.com]
> Sent: Monday, April 02, 2001 2:31 PM
> To: Barbir, Abbie [CAR:CC70:EXCH]
> Cc: Oliver Spatscheck; cdn@ops.ietf.org
> Subject: RE: hard questions: request routing
>
>
> OK. So you are constraining the topology at some early stage.
> This takes
> dynamic topology out of the scope of the RR protocol. This is
> definitely
> one solution. I assume you restrict the routes (which might form
> a graph) to a tree using the DPS. Is that correct?
>

I am not really constraining the topology. The algorithm works for an infinite number of
hops. The whole point of negotating the content hops is to give the content provider the
ability to control and track his content.

The technique defines the content tree. The DPS system is only used as a trigger
to point to a new level in the tree. It could also be used to shrink the tree
for example, when content is deleted.


> Again you make the topology completely static, this is one solution
> to the problem, however, it deals not very well with CDN failures.
> You could argue that this is a rare case... .
>

The topology is negotated. Keep in mind this applicabale for a certain content, not the
whole CDN content.

Regarding CDN failure, here you are mixing the issue again. The failures are solved by the
metrics , not the content tree.!!!!!!!!!!!!!!


> So you have three trees in your solution:
>
> - content tree
> - redirection tree
> - content routing tree
> You defined the first two but not the last. To help me not to confuse
> them a definition would help.
>


I only have one tree, basically the content tree (sometimes i call content routing tree)

The redirection tree may also be called the route computation. This is properitory.

> Overall your solution moved the problem of loop detection and
> avoidance from RRS to the DPS. You further argue that updates to
> the DPS are rare and therefore, a global atomic update is feasible.

The role of the DPS is just to advertize that new content has been added to the tree.
The DPS do not play any role in the loop prevention problem.
The DPS only trigger the event

>
> Does that summarize your solution correctly?
>
you got a little bit closer, but again start thinking in terms of content hops as
opposed to network hops.


> Sorry if I am slow in understanding your solution. However, it is hard
> to understand a solution having different scenarios in mind.
>
> Oliver


oliver, do not apologize, you are a genius, and i do appreciate all of your efforts
and feedback. you have given me a good run for my money.



abbie