OpenFlow Comes to Bat

7/5/2011

3:50 PM It's OK to be excited about the OpenFlow specification for switching, even if you have no idea what you would use it for

COMMENTS

Newest First | Oldest First | Threaded View

Add Comment

rmota1234 12/5/2012 | 5:00:14 PM

re: OpenFlow Comes to Bat

Hi Craig...Great baseball analogy on OpenFlow.  It's well done and gets the point across, I agree with your assessment and are seeing the same things. 

It will be interesting to see where it goes

Ray

Reply | Post Message | MESSAGES LIST | START A BOARD

Pete Baldwin 12/5/2012 | 5:00:03 PM

re: OpenFlow Comes to Bat

Wellll... I'm not sure that's what I meant.  After all, OpenFlow started as a very specific solution to a problem: Researchers wanting to experiment with new types of network control.

> I'm not clear what problems it solves for enterprise and network operator customers.

That's closer to what I meant. On the vendor side, there's a lot of excitement for OpenFlow but few concrete plans, AFAIK.

But the interest doesn't seem to be defensive. That is, I don't think vendors are investigating OpenFlow because they're afraid of being left behind. They think there's some gold here worth mining.

Reply | Post Message | MESSAGES LIST | START A BOARD

Duh! 12/5/2012 | 5:00:03 PM

re: OpenFlow Comes to Bat

you're saying it's a solution looking for a problem.  

I'm not sure I completely agree with that.  It can be a useful tool for conducting large scale experiments on research networks like GENI, Internet2 or their equivalents in Europe and Asia.    I'm not clear what problems it solves for enterprise and network operator customers.   And off the top of my head, I can't see a way to use it to address any of biggest architectural problems I see with the Internet.

Reply | Post Message | MESSAGES LIST | START A BOARD