For the past few years, SDN talk has centered around the packet layer and the data center use case, but that focus is now expanding. Just this April, the Open Networking Foundation (ONF) created the Optical Transport Working Group, which will develop SDN strategies for optical technologies, including optical transport networking (OTN) and photonics, as well as packet-optical integration. Packet-optical integration in the controller enables a new advanced capability referred to as multi-layer SDN.
Why the need for multi-layer SDN?
The networking industry has attempted to bring multiple layers into a consistent paradigm before, but has essentially failed at enabling the benefits of multi-layer integration. SDN's logically centralized network intelligence and ability to leverage cloud computing for almost unlimited compute power enables it to evaluate all layers of the network concurrently to determine where best to send traffic. Today, a service is typically transported at a single layer. With multi-layer SDN, a network can transport services over the most efficient technology, not just the predefined transport technology. For example, a switched Ethernet service could traverse the network partially over OTN and then over Multiprotocol Label Switching (MPLS) before being handed off as Ethernet, if that were most efficient.
Additionally, if bandwidth at a particular layer is exhausted in some portion of the network, multi-layer SDN can evaluate options and dynamically add bandwidth from a lower layer or reroute traffic from upper layers around the point of congestion. In most networks today, the ability to alleviate congestion by obtaining bandwidth from a lower layer takes days, weeks or months as the administrators for each layer must negotiate and execute the new capacity request through multiple email an network management system (NMS) screens. Intelligent multi-layer SDN can free the network and its operations team from this work and make the best-path determination, weighing layer tradeoffs, in milliseconds.
Multi-layer SDN: Congestion management and network optimization
In addition, multi-layer SDN networks can eliminate the need for hold-down timers. Hold-down timers are provisioned waiting periods upper layers use to enable lower layers to react to failures before the upper layer. Instead, the SDN controller will simply address the failure immediately at the most appropriate layer, which will result in shorter downtimes.
Further, applications could be written that provide automated congestion management, dynamic pricing and network optimization.
Automated congestion management recognizes prolonged congestion in the network and works with the controller to add bandwidth at a lower layer, to alleviate the congestion, or asks an upper layer to reroute traffic around the congestion.
Dynamic pricing analyzes a service request in the context of current and predicted service demand, and network resource supply, at multiple network layers and offers different service options at different pricing levels. This helps to maximize revenue per connection and to minimize resources consumed by incenting customers to select off-peak periods.
Network optimization evaluates network resource use at all layers and recommends or executes service path changes while ensuring SLA compliance. It thus makes more effective use of the network, and relieves current or potential congestion
What will make multi-layer SDN a reality?
Before these applications can be used, the network needs the three new capabilities to make basic multi-layer networking a reality: multi-layer topology; multi-layer cost paradigm; and multi-layer path computation. First, the controller must be able to pull together topologies of multiple layers and understand how they can interwork throughout the network. The controller must also establish and understand the "cost" of utilizing resources within and between layers of the network in order to evaluate tradeoffs between different paths across the network. Administrative cost, CAPEX cost, operational cost and other factors may be used to calculate this "cost." Finally, the controller must be able to map the resource costs into the multi-layer topology, determine the set of viable paths in the network and the network equipment that meet the service SLA, and determine the best path for that service.
With the inexhaustible processing power of cloud computing, executing these functions in an SDN-powered Wide Area Network (WAN) is achievable, even with the need to do many thousands of path computations per second. Service providers and enterprise reap the rewards of a network that uses less bandwidth (in the range of 15% to 60% less), is more automated to reduce operational expenses, and is more agile and allows for new dynamic services. Multi-layer SDN also makes the network more profitable, due to dynamic pricing for dynamic new services that providers and enterprises can program on their own.
Multi-layer SDN, particularly in the WAN, is poised to be one of the few technologies that can help service providers and enterprises contain their equipment and operational budgets in today's highly connected world, where users have a bandwidth appetite that is growing more and more insatiable while expecting low prices. Multi-layer SDN in the WAN will be critical for network providers and large enterprises to deliver services dynamically with maximum efficiency in today's highly competitive marketplace.
About the author: Rob Tomkins is senior advisor for portfolio strategy in the office of the CTO at Ciena.