SDN Use Cases: Disaster Recovery

Disaster recovery has always been critical. Staying competitive means having resilient networks that reliably meet customer SLAs. Yet the complexity of running today’s service provider networks makes it more challenging then ever to consistently deliver excellent service.

Exacerbating this difficulty are regulations such as the upcoming General Data Protection Regulation (GDPR). This law, which goes into effect on May 25, 2018, has specific disaster recovery requirements for any business that has customers or employees in the EU.

SDN for Disaster Recovery

SDN can automate many DR requirements across the WAN via functions such as path provisioning and traffic engineering. SDN also provides the opportunity for service providers to add new failure recovery services. For example, because enterprise data is now stored in multiple places, WAN performance becomes more critical. The quality of the user experience directly impacts enterprise productivity. Service providers must ensure the same experience after an event whether these employees access applications and data over the intranet or Internet.

Disaster recovery is also a case for multi-layer optimization across SDN. Many service providers want to use SDN technology to converge the management and orchestration of both the IP/MPLS and optical transport network layers. For example, if the network suddenly loses an optical link, the management system must know this in order to route traffic across another [IP] path. It must have the intelligence to determine if that path is acceptable and configure it correctly.

This intelligence must come from a real-time and historical understanding of traffic volume and patterns as well as how traffic should be routed in different failure scenarios. SDN controllers must have a source of this intelligence – path computation and optimization technology – to instruct them how best to re-route traffic in these situations as well as ensure that the alternative path will not impact other applications and services.

Recent Implementations

We recently employed our SDN management intelligence for a European mobile operator in a multi-layer optimization situation (outlined by Packet Design CTO Cengiz Alaettinoglu in this blog post). Working with our SDN integration partner Ciena Blue Planet, we were able to reroute service traffic away from network links that had CRC errors – before the service traffic experienced significant degradation.

This was accomplished with our Explorer SDN Path Provisioning application, which automates and provisions service path computation. It uses the real-time IGP/BGP routing telemetry, traffic matrices, policy-based path computation and optimization engine, and interface to SDN controllers provided by the Packet Design Explorer SDN Platform. Speaking of adding new disaster recovery services, providers can also use the app to offer a diverse path service for customers who must ensure there is no single point of failure.

Another Packet Design customer, XL Axiata, one of the largest telecommunications providers in Asia, is using our Explorer SDN Traffic Engineering (SDN-TE) application to build in network resiliency for its 40+ million customers. The service provider experiences link failures often – on both sea and land – because of cuts to fiber optic cables on the ocean floor and a high rate of construction damaging equipment across Indonesia.

The SDN-TE app re-balances network loads automatically – often an arduous and lengthy manual process for network engineers. Like the Path Provisioning app, it is fed by the Explorer SDN Platform’s unique set of real-time, historical, and predictive analytics. Using these, the app calculates and provisions traffic engineered tunnels to shift load from congested links to less heavily used ones.

These are just a few of the examples of how disaster recovery is a valuable use for SDN. Service providers who automate it under various scenarios can increase business agility and create differentiation in competitive markets.