The research-and-education (R&E) space has been among the most fervent areas of networking innovation in the last decade.

Consider the collaboration, medical and scientific capabilities enabled by networks such as Texas’s Lonestar Education and Research Network (LEARN), Washington’s Northwest Open Access Network (NoaNet) and Pennsylvania’s KINBER (Keystone Initiative for Network-based Education and Research) in the United States; BCNET in British Columbia, Canada, and LightNet in Italy’s Trieste territory. High-definition broadcasts of surgical operations, remote diagnosis, precise and remote control of microscopes and other scientific instruments, high-speed downloads of entire libraries of data, distance learning, real-time collaboration and grid computing are among the cutting-edge applications carried across the world’s R&E infrastructures.

But it is not just the students at the networks’ member institutions that have something to learn here. The R&E networks are offering some valuable lessons for enterprise and carrier network operators as well. Modes of network operation are predicted to emerge as increasingly popular among enterprises and carriers in the coming years in R&E infrastructures globally.

Control-plane Interoperability

 Packet-optical integration—undergirded by seamless interoperability between the Internet Protocol/Multi-Protocol Label Switching (IP/MPLS) and transport layers—is a prime example of the lasting impact that R&E is likely to have on carrier and enterprise networking.

In the network model adopted widely across R&E infrastructures, software-programmable transceivers, reconfigurable optical add/drop multiplexing (ROADM (News - Alert)) technology and Generalized MPLS (GMPLS) control planes work in sync with one another to allow for automated, end-to-end inventory, discovery, reservation and setup of optical-fiber wavelengths. The optical transport domain is encapsulated in a virtual topology that exposes available network resources to the packet domain, and the GMPLS control plane-based service manager maintains up-to-date resource availability. With only mouse click on a router’s graphical user interface (GUI), optical wavelengths can be set up across the Dense Wavelength Division Multiplexing (DWDM) network for real-time resource sharing between R&E institutions engaged in some type of collaborative application for distance learning, telemedicine, etc. The IP/MPLS routers and transport equipment along the path between network destinations are automatically configured, as the service manager orchestrates connection management via hundreds of control messages across appropriate network elements.

In this way, via service management enabled by the GMPLS control plane, bandwidth can be dynamically provisioned between any two points on a network for a given period of time. It is a breakthrough capability with tremendous value beyond the packet optical networks (PONs) employed in the R&E space.

Value Beyond R&E Networks

 In such a model of packet-optical integration, the command to turn on and off wavelengths and to turn up and tear down circuits so quickly and easily is put in the hands of information technology (IT) staff using their familiar and proven control and management mechanisms. What does such a capability mean for carrier and enterprise network operators?

By integrating packet and optical domains into one seamless network, a network operator’s service velocity, efficiency and reliability are dramatically increased. Time to market for new services is shaved from months to minutes, as expensive truck rolls to configure equipment along network paths are eliminated. Network resources are more efficiently utilized, as bandwidth can be dynamically shared as necessitated by demands, events or typical usage patters—as opposed to overbuilding links to accommodate potential surges in bandwidth demands.

These are timely capabilities for the business environment in which enterprise and carrier network operators find themselves today. The days are gone when companies are willing to install giant network pipes, pay for them year-around, and use them to their fullest capacity only when they are needed.

Network operators must be nimble enough to dynamically and cost-effectively provide their users with robust, sophisticated capabilities on demand—and reassign such resources to other users just as quickly. Packet-optical integration—enabled by control-plane interoperability—can convey such a high degree of flexibility to network operators today.

Conclusion

 Control-plane interoperability between the IP/MPLS and optical transport layers is a revolutionary capability. With a holistic view of all network elements and the ability to execute commands across multiple domains, network operators are able to advance into automated, dynamic provisioning that dramatically enhances service velocity, network utilization and profitability. Already proven in R&E networks around the world, it is a capability that figures to have more and more impact on carrier and enterprise infrastructures in the years ahead.