[Sdnp] Proposed SDN BOF charter

[Thomas Nadeau]

On Sep 21, 2011, at 9:13 AM, Anton Ivanov wrote:

> [snip]
> 
>>> There is a bit of a misunderstanding - I want to put the foundation in order to be able to figure out the inter-domain case. I want to have that foundation day one, but I have no intention of figuring out the case itself day one.
>> 
>>  I think that is achievable insofar as we will need to define HA capabilities from day 1, which requires SDN Conductor-to-SDN Conductor communication.
> 
> Hmm... That is an interesting one...
> 
> You probably need that one if you want to have different Conductors (and different implementations) interoperating as a part of the same admin domain and commanding the same devices. In that case needing a Conductor-to-Conductor protocol is the easiest option.
> 
> You can however have HA without a standardized SDNC-to-SDNC implementation and in fact that is way easier to implement (just snapshot database regularly from master to slave or use HA database in the back with appropriate locking). That however automatically makes the case of "multiple different conductor implementations"in the same admin domain nearly impossible.
> 
> It does not prohibit multidomain though. 
> 
> In a multidomain instead of using SDNC-to-SDNC (which you can also use for HA) you use SDNC talking as an SDN Client to another SDNC (aka the "russian dolls"). 
> 
> The latter will make adoption and implementation easier. One less critical thing to implement. All you need to specify is the client protocol and how it works internally is "none of our business".
> 
> The former is neater and better in the long run.
> 
> Both have their advantages...

	The internal model makes a lot of sense in terms of speed of implementation; however, that does not facilitate deploying SDNCs from multiple vendors that can handle HA.  That is why I think we need some sort of HA/sync protocol.

	--Tom


> -- 
>     Humans are allergic to change. They love to say, "We've always 
> done it this way." I try to fight that. That's why I have a clock 
> on my wall that runs counter-clockwise.   -- R.A. Grace Hopper
> 
> A. R. Ivanov
> E-mail:  anton.ivanov at kot-begemot.co.uk

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lucidvision.com/pipermail/sdnp/attachments/20110921/73ef41ec/attachment.htm>