This vendor-written tech primer has been edited by Network World to eliminate product promotion, but readers should note it will likely favor the submitter's approach.
Although much of the Software Defined Networking discussion to date has been about the data center, software-defined WANs (SDW) also show great promise. In essence, an SDW enables IT teams to manage network services by decoupling the system that makes decisions about where traffic is sent (the control plane) from the underlying systems that forward traffic to the selected destination (the data plane). This can deliver greater simplicity, security and resiliency, which in turn can lead to lower operational expense (OpEx) and capital expense (CapEx).
While high-quality SDW solutions certainly aren't "cure-alls" for the myriad of cost-of-change issues enterprise IT and purchasing teams routinely face these days, they are certainly options with the potential for positive change.
Exactly how can a high-end SDW solution help? Here are a few examples:
* Simplicity. A great deal of IT spending, of course, goes toward getting new things deployed into existing environments. From an OpEx perspective, the main emphasis is usually on day-to-day maintenance, modifications and other related tasks. And any time you make a modification that involves complex firewall rules, VPNs or integrating with other sites with traditional approaches, things can get very costly very quickly. And, risk your networks security.
In contrast, SDN solutions can simplify the tasks of provisioning and maintaining network resources, which not only lowers the cost of network management but also reduces the need to constantly reconfigure networks, a process that's time-consuming and prone to human error. From a CapEx perspective, the cost of scaling -- from buying new hardware to expanding server rooms -- can also be very expensive. Rather than taking this route, SDWs offer IT managers the chance to be more scalable at the software level. So, instead of constantly buying new appliances and expanding server rooms, organizations simply purchase new software licenses.
* Security. Today, a typical approach when integrating two networks is to grant universal access first and then to figure out which connections need to be secured and how and when to do it. This affects both OpEx and CapEx in different ways. From the OpEx point of view, the process of securing the system can be both costly and complicated. IT teams have to get VPN concentrators, figure out all their firewall and access controls, add intrusion detection capabilities, and then monitor and manage all access by various user groups.
In contrast, SDW solutions enable IT teams to build point-to-point tunnels, isolate addresses spaces, and then secure them with certificates. What's being granted is extremely specific, so those teams can more easily protect system security. From a CapEx perspective, typical solutions usually involve buying numerous very specific vertical solutions to address every piece of the environment that might be vulnerable. Instead of requiring all this, SDW solutions approach the issue differently. They permit access only as desired right from the start through certain tunnels that are secured by certificates. This, again, is a much more specific -- and inherently more secure -- approach.
* Resiliency. Another way SDW solutions can help reduce costs on both the OpEx and CapEx fronts is by providing sophisticated traffic management and link-bonding capabilities to make networks far more flexible, responsive and resilient. By using built-in traffic engineering algorithms, for example, IT administrators can prioritize traffic by application, protocol or port. Alternatively, by employing store and forward features, they can assure that all data gets to the proper destinations while the network also optimizes bandwidth utilization and delivers on quality-of-service benchmarks. This, of course, makes for a far more efficient -- and cost-effective -- network that requires less operational and capital outlay.
Advanced SDW approaches might not be able to solve all the critical problems faced by IT and capital-equipment purchasing teams. But they do offer benefits that, in addition to increasing network performance, can mean considerable cost relief for IT teams constantly pressured to "do more with less" and for enterprise purchasing teams constantly pressured to "get more for less." For most organizations, these approaches offer a way to turn network resource management on its head -- something I believe will soon win much wider acceptance.
Jay Friedman is president of Distrix, based in Vancouver, B.C., Canada