Automating the intelligent transport network

Staff writer
24 Mar 2014

Mike Capuano, Infinera's VP of marketing, explains why comprehensive automation is a critical foundation - "you can't software - control something if it requires a truck roll or a manual patch"

Optical Evolution: What is Infinera's approach to enabling telcos to cope with soaring bandwidth demand?

Capuano: With the continued rise of cloud and real-time entertainment services, we all know that service providers are facing the perennial challenge of continued network investment to meet massive bandwidth increases in the face of decreasing revenue per bit.

Our transport system is quite different compared to conventional systems for three reasons. First, it is based on highly integrated and scalable PICs. Second, it is a purpose-built converged DWDM and OTN switching platform, and third it is completely automated via a unified GMPLS control plane at both the OTN switching layer 1 and the optical ROADM layer 0.

You are a Silicon Valley based company with strong expertise in large-scale PICs. How important is that in developing optical transport for service providers?

Manufacturing large-scale photonic integrated circuits (PICs) is our foundational innovation. We do not sell the PIC as a component, but use it as a key building block for our Intelligent Transport Network platforms.

With the DTN-X for example, which we started shipping in 2012, the PIC enables us to deliver single card 500 Gbps long-haul FlexCoherent super-channels vs conventional discrete component-based techniques, which can only support 100G per line card today. This level of integration of the coherent optical engine in turn provides the power, thermal and space headroom to integrate 5 Tbps of non-blocking OTN switching at the system level into a single chassis with forward scale to 12 Tbps per chassis and 240 Tbps in multi-bay configurations.

Because our solution is purpose-built for this convergence, we can configure the system with any mix of DWDM super-channel line-side and tributary client-side capacity with "no compromise", always maintaining 5 Tbps of usable capacity per chassis, which is unique in the industry.

Tell us about your Automate Everything announcement.

Our Intelligent Transport Network offers full automation. For example, install two 500G super-channel cards, one at each end of the route, and the system activates them automatically either at a full 500G of capacity or Nx100G using the industry's only "Instant Bandwidth" capability. Services are provisioned using a WYSIWYG GUI from 1G to 100G through the OTN switch, which is used to dynamically map these services to a pool of OTN containers so that any service can be carried without any restriction into any wavelength speed or type - we call that Bandwidth Virtualization.

The switch and GMPLS also support FastSMP - the industry's only 50-ms hardware accelerated shared mesh protection. Once theses digital services are packed into the super-channels the same GUI and GMPLS control plane supports point and click routing of optical super-channels through the industry's first super-channel CDC FlexROADM - and these new optical switching capabilities are the key elements of our recent Automate Everything announcement. The automation enables service providers to deploy services faster so they can be more competitive, use their network more efficiently for lower capex and automate previously manual tasks for lower opex.

Related content

No Comments Yet! Be the first to share what you think!