Wednesday, December 12, 2012

Carrier Ethernet OAM Part 1

Service providers are determining where there is a need for more fiber and what kind of reach it can go to the rural communities. As more network elements are deployed to keep up with bandwidth demand, so is there an increased importance in scale for network management and monitoring performance.

First step is data collection. If you can't see it, you can't manage and control it. Beyond the Carrier Ethernet NOC, questions are being asked. What do customers want? What do service providers want? There is a growing need for speed for consumers and enterprises. We are seeing incremental increases in bandwidth speed all the time. 

A little historical perspective: Remember when the Hayes 9600 bit modem put the internet in the hands of the masses? Remember when a T1 was thousands per month? Speeds increase, costs come down. It's a classic case of economics and technology innovation.

Service providers want to see their costs go down. As bandwidth demand increases, their revenue is not moving in a parallel line to it. As customers see the advantages of higher speed, the service providers want to see the money. A bigger pipe just gets you so far.

Are you willing to pay for higher SLAs? Yes, enterprises are asking for it. Willing to pay extra for security? Certainly the government and military demand and pay for it. It would make sense that enterprises with sensitive information would pay for extra security. Are you willing to pay for higher quality or a class of service? Sure, but only if there is a portal for customers to see their service usage stats, performance metrics and can provision for their needs.

Back to the Carrier Ethernet NOC story. Controlling, measuring and reporting Ethernet service in a standards-based way across services and across vendors is a key to helping service providers with business continuity and reducing OPEX. Although the MEF-Ethernet management model has an established baseline, not all of the Carrier Ethernet vendors use standard MIBs and implement their own RFCs to support OAM and CFM by querying custom CLI command sets. Every service provider has a hodgepodge of systems that do different functions. That's the way it is, by design, best of breed or by legacy of investment. There are two ways to go about this, a unified system or an integrated approach.

No comments:

Post a Comment