On Sep 9, 2012, at 2:48 PM, Curtis Villamizar wrote:

> The question for the WG is whether this work is of any interest to the
> IETF.  My opinion is a *NO*, there is no interest in this work.
> 
> At this point we need to hear other opinions, but so far I hear no
> support.


I have an interest in this area, but like Curtis, would like to suggest that we 
need to understand the degrees of freedom that are possible and practical.  We 
should try to focus our directions on the pragmatic.

Like Curtis, I'm highly skeptical of the ability or opportunity for a provider 
to shut down an entire system.  If that were truly practical, it would imply 
that the system would be wholly unnecessary for primary or protect paths.  
While there is a substantial optimization possible because of diurnal traffic 
patterns, I suspect that the granularity of the power savings is at the link 
level, not at the system level.

Unlike Curtis, I'm not convinced that manipulating DWDM optics on the fly is an 
insurmountable challenge.  I agree that it is non-trivial and certainly not 
something supportable immediately, but as our control systems mature, it is not 
inconceivable that we will someday get to the point where this can be automated.

Thus, the net is that the real power savings that are possible today are at the 
NPU (and associated fabric) level.  Having a protocol to support this, as well 
as the TE facilities to automate the process seem like a worthwhile direction.  
And, as always, solving problems at the intra-AS level is far, far easier than 
addressing them at the inter-AS level.

Regards,
Tony

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

Reply via email to