That is a useful guideline, but I do not think it should be a strict
requirement. Right now, the only dependency is between a cost map & a
network map. But we have talked about other relationships, such as topology
or routing maps between PIDs. So I think the SSE update draft should be
general enough that it could be applied to new dependency relationships, and
those future relationships might result in more general graphs than the
rooted-tree of a network map and its cost maps.

Section 6.5 does recommend that the resource set for each update-stream
resource be closed under the dependency relationship. However, that allows a
server to define one update-stream for several network maps & their
dependent cost maps, while I think you are suggesting that each network map
(and its dependent cost maps) should have its own update-stream resource. We
could extend the recommendation by adding, ³and no subset  of the resource
set should be closed², which I think accomplishes what you suggested. But I
do not see how that would simplify either clients or servers.

- Wendy Roome

From:  EXT Gao Kai <[email protected]>
Date:  Wed, March 9, 2016 at 01:05
To:  Wendy Roome <[email protected]>, <[email protected]>
Subject:  Re: [alto] State of the WG

On 08/03/16 22:14, Wendy Roome wrote:
> * Delivering push-updates for multiple resources via one stream was only
> partly to minimize the number of streams. The bigger reason is that cost maps
> depend on network maps, and if the network map changes, the cost map changes
> as well. It is easier for the client to handle those dependencies if the
> server delivers updates for all those resources via a single stream, in a
> predictable order.
Maybe we should make it explicit that only the cost map with its dependent
network map(s) can be pushed in one stream?
> 


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

Reply via email to