ADN and SDN: a win-win partnership
The 2013 Open Networking Summit (April 15 to 17) showcased the immense innovation that has taken place in the networking industry in the past year. Startups and incumbents alike have developed significant offerings in SDN controllers, network virtualization, network function virtualization, software appliances, and the like.
Among the announcements and demonstrations were: the OpenDaylight project, Intel’s Open Network Platform, Cisco’s Open Network Environment, and many others. Together with these there were also significant end user initiatives and proofs of concept including Goldman Sachs’ in-house initiatives.
However, a couple of things struck me after 3 days of sessions:
- We were no closer to having a universally agreed to definition of SDN
- There was a big blur on the overlaps and differences between SDN, Network Function Virtualization, and Network Virtualization
One way of looking at things is from the viewpoint of the applications that ultimately use the network. In previous blog posts, our CEO, Pascale Vicat-Blanc, has introduced the concept of Application Defined Networking (ADN) in Clouds. ADN is compelling because in a lot of respects it is the end user view of where networking needs to be to keep up with the Cloud and DevOps revolutions that we are currently living through.
Networks clearly need to keep up with the tremendous agility and rapid evolution that we see in cloud-based applications today. Looking at it this way, one requirement clearly stands out, Networks need to be Programmable.
The separation of the control and data planes, centralization of the control plane in a controller (or federation of controllers), and a Network Operating System that handles the low-level configuration of physical and virtual network hardware would all greatly enable the programmability of the network.
If we can call this plumbing SDN, then SDN is also one of the enablers for network virtualization and network function virtualization.
ADN prescribes a tight real-time feedback loop between the network and the application. This is beyond than the network being application-aware or the application being network-aware. This two-way interaction enables the application and its (virtual) network to adapt to each other in real-time. Like “breathing” together. For example, a network can learn the bandwidth and latency requirements of communication flows emanating from an application and can adapt itself to enforce these constraints end-to-end for the application. The network can re-route these flows in response to bottlenecks in order to maintain the application SLAs. On the other hand, if the network is unable to resolve bottlenecks that arise, then the application can adapt itself to workaround these issues by moving workloads, adding compute capacity, etc.
ADN is all about enhancing the agility of applications and the elasticity of networks. It is in keeping with the spirit of DevOps – Communication, Collaboration and Integration (albeit between the application and the network vs. software developers and IT professionals).
From the above example, it is clear that there is a win-win partnership between ADN and SDN. They are complementary in nature. SDN makes the network programmable and ADN provides an actual use-case for programming the network.
At the ONS, venture capitalists exhorted startups to focus on the business value of their solutions. Lyatiss’ pioneering of ADN is driven entirely by business value. ADN+SDN taken together, provide a revolutionary approach to managing applications in the cloud. They increase the efficiency of cloud operations by automating well-understood and low-level operations functions. They also make applications more robust, enable them to perform better, and make them cheaper to develop and operate.