So, NetworkWorld ran a story on a solution from startup Pertino’s Network as a Service (NaaS). It actually sounds like a pretty cool solution that’s very simple to setup. The backbone of the solution uses a SDN based control plan that runs on AWS. This solution is as Cloud as Cloud gets. However, this isn’t what I picture when I think of NaaS.
I’ve always pictured NaaS as being the ultimate marriage between SDN and Cloud computing. I pictured a scenario where you’d have a hyper-fast connection to a telco network. I’m thinking something along the lines of a Google Fiber connection to a home or business. This part of the solution is the only manual provisioning that would need to be done. You have to have some type of access layer after all. From this point forward everything else would be managed by next generation SDN that really doesn’t exist today.
You would be able to go to a NaaS provider which could be a completely virtual company like Pertino that has for a lack of a better term SDN peering relationships with all of the backbone providers. You’d be able to build on demand networks between as many sites as needed with a speed and class of service based on your application needs and budgets. This service could even have an API that builds and tears down networks like AWS instances. The key in my approach is the high bandwidth access layer provided by Google or whatever last mile provider you have available.
From a billing perspective, you would pay an access charge to this local telco but your NaaS provider would handle all of the intermediary charges based on your class of service. You could in theory have multiple NaaS providers our switch NaaS providers one day to the next. Some form of SDN has to be in play to make all of this work as seamless as it sounds but to me this is the true promise of SDN and NaaS.
I’d be interesting in what you’d consider as NaaS or even what you think of my interpretation of NaaS and SDN.