As recently announced, Bell Labs researchers have uncovered potential cost benefits of using a converged operation that combines routing and optical transport into an integrated multilayer network model.
Its study shows that:
While this is pretty cool for Tier 1 and 2 network operators, I couldn't help but wonder what this means for everyone else.
Arnold Jansen, a senior product manager at Bell Labs' parent Alcatel-Lucent, allowed me to pick his brain about this innovation. He threw out some interesting, compelling, and important factoids that provide insight into our public network.
For example, Arnold told me that VoIP traffic must be restored within 50 milliseconds. If you think about it, this is not a lot of time -- especially should a wavelength "fail," since recovery time for such an incident is about 10 seconds. So, in the scheme of things, the life of that VoIP traffic expires after 50 ms of disruption. This kind of failure has significant impact on VoIP and other real-time traffic that rides the network.
The Bell Labs technology combines optics with the IP Fast Reroute protocol over a GMPLS network. Cisco describes the role of the fast-rerouting function as such: "Fast re-routing protocol Fast Reroute (FRR) is a mechanism for protecting MPLS traffic engineering (TE) LSPs (Layered Service Provider) from link and node failures by locally repairing the LSPs at the point of failure, allowing data to continue to flow on them while their headend routers attempt to establish new end-to-end LSPs to replace them. FRR locally repairs the protected LSPs by rerouting them over backup tunnels that bypass failed links or node."
Existing 1+1 optical networks retain 50% of network capacity in reserve; this is where the Bell Labs technology kicks in, opening up more network capacity. As mentioned, service providers would be able to increase available bandwidth by up to one third of deployed capacity by combining MPLS and a GMPLS UNI.
I've held for a long time that adding intelligence to the network will further the self-healing ideal. Consider current issues with keeping VPNs alive and network access across the enterprise always on and available, with traffic flowing 24x7. These demands are now obtainable with increasing levels of service and decreasing costs. Bell Labs has leveraged optics, routing, and software by way of convergence to bring down the high cost of retaining 50% reserve network capacity for resiliency.
How does all this benefit enterprise users onboard with network operators? According to Arnold, the cost per bit goes down each year by 8%. While consumption goes up exponentially, 40% is a marked improvement over existing network architecture.
But the greater benefit, I think, is improved resiliency for cloud providers such as Amazon Web Services. Moving to cloud services that incorporate both optics and routing will improve reliability of real-time traffic with added improvements centered on resiliency. Bandwidth consumption continues to rise but this doesn't mean that convergence is stagnant since enhancements such as this have significant overall impact on how networks should continue to improve and operate.
The network -- note that I did not say "Internet" -- requires innovation and constant improvement. It is a moving target, forever not static and always challenging.
Follow Matt Brunk on Twitter and Google+!
