[Sdnp] Proposed SDN BOF charter
ping at pingpan.org
Thu Sep 22 18:49:29 EDT 2011
On Thu, Sep 22, 2011 at 11:00 AM, Monique Morrow (mmorrow) <
mmorrow at cisco.com> wrote:
> Belated ack Tom, and a very reasonable cut to defining the BOF charter,
> just a small question:
> Do you seen any differentiation from the proposed work here to say that of
> the ONF Extensibility WG?
> TIA- Monique
To my knowledge, just in the data center area, there are many proprietary
solutions on this already. Let's face it, the name of the game in data
center infrastructure is to integrate computing, storage and networking as a
whole in the most efficient manner. Given its scale, it's impossible to have
a single vendor solution (except, perhaps, Google ;-)). As such, we need a
standard interface between different parts.
For IETF, our goal is to present the network interface to the applications,
such that network resources can be best utilized. (It's not our goal to
re-define VM operation though).
Further, it's been a myth to state that the high cost of network equipment
is due to vendor monopoly, and a new protocol would drive down the equipment
cost. On the contrary, adding new control protocols on equipment will
increase the cost, not reducing it. The driver behind network equipment
commoditization is Moore's Law, not new protocol or Chinese. ;-) Just take a
look at the yearly equipment cost vs. capacity chart on Ethernet and
ROADM... It's my belief that we have collectively developed a sophisticated
and efficient network. The challenge is in making it available for upper
applications to make the best use of.
I am open to standardize this in IETF, ONF, or open source code... I don't
carry a religion with me so long as this work can benefit the industry.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the SDNP