Each time it’s asked, I realize that it’s a great question that underscores the confusion that still exists in the marketplace despite our best intentions. The market (vendors and customers alike) is still trying to define SDN and gain clarity about how it can help solve significant IT issues.
SDN is one of the more confusing trends I have personally marketed my way through. It can be fun and invigorating on a good day and frustrating on the days when your conversations are dominated by vernacular and not focused on addressing customer pain. That pain is represented by the manual processes that are fundamental roadblocks to end-to-end, hands-free provisioning. Many of these companies indeed want the network to be part of an orchestration process. They have also maintained that they want to be able to move a virtual machine (VM) from one network domain to another.
Two different approaches appear to be emerging in addressing these pain points. The under-layer or physical network infrastructure approach is focused solely on the network fabric itself. You might deliver your solution on a white box or a legacy vendor with your own hardware design; nonetheless, you care about routing/switching/management and provisioning the network itself. You might support OpenFlow. And as I noted, these solutions are delivered on a switch. The ASIC in the switch is an integral part of the story.
The overlay/virtual network infrastructure approach is focused on building tunnels (also called overlays), and that aspect is called network virtualization. Building virtual network function capability into those tunnels is part of NFV. For example, you may want to connect two isolated domains without changing the underlay and possibly spin up a firewall on that tunnel. These solutions are delivered on an x86 server platform. There are no ASICs, and the rationale is that you can scale here through all of those multicore CPUs.
Many prospects are starting to understand the difference and the relevance between under-layer and over-layer segmentation. In some cases, the under-layer focus is what the market is calling SDN. Anything running on a server is either NV or NFV and therefore, not SDN. Hey, don’t blame the messenger!
No matter the approach, network agility is a goal everyone can agree upon. To get there, we must embrace software in networking similarly to how it’s evolved on the server side. If we all agree on that notion, businesses considering this must ask themselves how they want that spectacular solution served up to them?
Network virtualization and NFV are indeed software-led applications. Having said that, network virtualization and NFV are both delivered to a customer on a server. The association with the network is more in a logical sense only, in that you are controlling or interacting with the flow, or interconnecting logical domains.
And networks will always be delivered on a switch. That won’t change. Switches provide the scale to aggregate all those end devices through protocols that build the fabric that delivers the packets to those end stations.
As these software solutions come to market, it’s important let’s help the market understand if they are switch-based or server-based. It will really help us get to “the answer” faster.