As for your specific questions,
(1) Is this what¹s meant by the ³multiple switch model²?
I believe so, although this graph is a very simple example. My take is
that with the ³single switch² model, we think of every host on the planet
connected by a dedicated line to one central switch (with a LOT of ports).
That¹s the ³blob² model, where we show the net as a cloud with lots of
devices connected to it, and no idea about the blob¹s internal structure.
The ³multiple switch² model simply exposes some of the network blob¹s
internal structure.
(2) What sort of algebraic formulations do we expect .
That¹s messier, and I¹ll leave that to Richard Yang.
As for your pseudo-JSON description of the network, remember that nodes do
NOT have to be PIDs. I think all PIDs must be nodes, but not vice versa.
Also, I suggest using a separate ³node property service² for all but the
most essential attributes of nodes & links.
- Wendy Roome
On 01/27/2014, 15:00, "[email protected]" <[email protected]>
wrote:
>Message: 2
>Date: Mon, 27 Jan 2014 10:46:04 -0600
>From: "Vijay K. Gurbani" <[email protected]>
>To: "Y. Richard Yang" <[email protected]>
>Cc: alto <[email protected]>
>Subject: Re: [alto] Work items for re-chartering
>Message-ID: <[email protected]>
>Content-Type: text/plain; charset=ISO-8859-1; format=flowed
>
>On 01/25/2014 10:45 PM, Y. Richard Yang wrote:
>>How about give different milestones to different items, with some later
>>milestones? The chairs enforce more strictly the deadlines.
>
>For a recharter to be successful, we need to understand in sufficient
>detail what we want to do. The value proposition of the 7 items [0] is
>not under dispute, as is evident by the rather strong show of support
>thus far. (Reinaldo's reticence duly noted and appreciated.)
>
>Let's do a second order analysis on what we expect to have as an
>outcome of the work item, "An ALTO format for encoding graphs."
>Personally, I think that this is the most intriguing of the lot.
>
>So, in order to ensure a common understanding such that we are
>forearmed with answers to the questions we know we will get when
>we recharter and assign milestones, let's do a bit of armchair
>analysis on the encoding format for graphs.
>
>Consider the following ALTO triangle:
>
>
> PID-1
> / \
> 2/ \ 3
> / \
> PID-2--------PID-3
> 6
>
>Today, ALTO knows the pair-wise cost (i.e., {PID-1,PID2}, {PID-2,PID-3}
>and {PID-1,PID-3}) from the cost map. So if a resource consumer is in
>PID-2 and there are two resource producers, one in PID-1 and the second
>in PID-3, ALTO will choose the producer in PID-1 because of the low
>cost to get to PID-1.
>
>The network map will be encoded as:
>
> "network-map" : {
> "PID3" : { "ipv4" : [ "0.0.0.0/0" ] },
> "PID1" : { "ipv4" : [ "192.0.2.0/24", "198.51.100.0/24" ] },
> "PID2" : { "ipv4" : [ "192.0.2.0/25", "192.0.2.128/25" ] }
> }
>
>and a cost map as:
>
> "cost-map" : {
> "PID1": { "PID1": 1, "PID2": 2, "PID3": 3 },
> "PID2": { "PID1": 2, "PID2": 1, "PID3": 6 },
> "PID3": { "PID1": 3, "PID2": 6 }
> }
>
>Here, by design of the current ALTO protocol, the underlying network
>(switches, links) are abstracted behind PIDs.
>
>Issue 7 of the proposed rechartering effort [0] deals with a suitable
>format for encoding graphs in ALTO. My impression is that there
>are two distinct capabilities we are looking for: graph
>representation and graph transformation. The representation part
>appears easy enough (I think).
>
>Taking some liberties with the syntax, a coarse JSON GML encoding
>can be approximated as follows:
>
>graph [
> node [
> id 0
> label "PID-1"
> address-type "ipv4"
> hosts ""192.0.2.0/24, 198.51.100.0/24"
> ]
> node [
> id 1
> label "PID-2"
> address-type "ipv4"
> hosts ""192.0.2.0/25, 192.0.2.128/25"
> ]
> node [
> id 2
> label "PID-3"
> address-type "ipv4"
> hosts "0.0.0.0/0"
> ]
> edge [
> source 0
> target 1
> value 2
> type "undirected"
> ]
> edge [
> source 0
> target 2
> value 3
> type "undirected"
> ]
> edge [
> source 1
> target 2
> value 6
> type "undirected"
> ]
>]
>
>Two questions:
>
>1) Is the encoding above close to what we have been calling the
> "multiple-switch" model? Is something like this encoding one possible
> outcome of a work item related to Issue 7?
>2) The harder part is graph transformation. What sort of algebraic
> formulations do we expect to turn into protocol primitives that will
> allow us to transform graphs as is outlined in Section 4 of [1]?
>
>[0] http://www.ietf.org/mail-archive/web/alto/current/msg02345.html
>[1] http://tools.ietf.org/html/draft-yang-alto-topology-00
>
>Thanks!
>
>- vijay
>--
>Vijay K. Gurbani, Bell Laboratories, Alcatel-Lucent
>1960 Lucent Lane, Rm. 9C-533, Naperville, Illinois 60563 (USA)
>Email: vkg@{bell-labs.com,acm.org} / [email protected]
>Web: http://ect.bell-labs.com/who/vkg/ | Calendar: http://goo.gl/x3Ogq
_______________________________________________
alto mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/alto
