Is SDN going to die? See what everyone is saying!

With the advent of network automation, programmable networks, and software-defined networks, it’s time to take another look at where SDN fits in. SDN as a networking industry “buzzword” seems to have happily faded. But SDN still exists in other ways—in the data center or in the next-generation wide area network.

[[252873]]

Here are three analysts' comments on the development and future of SDN. Will data center SDN be the end all of technology?

Is SDN dead, or has it been called something else?

Lee Doyle, senior analyst at Doyle Research: In the data center environment, it is still called SDN. But it has changed in the face of players such as Cisco, VMware and white box network operating systems.

John Fruehe, independent analyst: It has changed. SDN started out in large cloud data centers, and companies like Google, Amazon and Facebook are using it as a transformational way to automate their networks. Over time, we've seen SDN evolve in different product lines. We've seen configurations that support multiple controllers, rather than a single appliance configuration. But a single network driven by a single console will probably never happen because there are so many SDN products on the market today. Overall, I don't expect data center SDN to go away.

Tom Nolle, president of CIMI: The problem with SDN is that the concept was hyped before it was adopted. By the time things really started to happen, it was old news. SDN may never be applied in the data center at the expected speed, and now SDN is in the stage of self-adjustment and mutual running-in with the business.

What has surprised you most about the evolution of SDN in the data center?

Doyle: I would answer this question with the key things that are driving the development of SDN. Everyone is focusing on agility, customer experience, and wanting to make private cloud as good as public cloud, which are huge drivers of SDN. Traditional networks still require too much manual intervention, and this is what we are trying to solve.

Fruehe: The most surprising thing is that there is a clear use case in the small business space. Companies like Ubiquiti, Aerohive, and Hewlett Packard Enterprise are launching products targeted at small businesses that give them a unified console -- a single console for managing wired and wireless infrastructure, as well as a host of other devices. This is an interesting use case because in a small enough area, one console can effectively control everything.

Nolle: Actually, no. When I did my initial research and predictions on this, I thought about the success of SDN in the enterprise, mainly in the speed of virtualization and hyperscale data center deployments. There is nothing new about SDN in terms of technology adoption or otherwise.

What's missing from SDN?

Doyle: There is no clear architecture or choice for users when it comes to software-defined data center networking. If you decide to go pure Cisco, you may only be aware of Cisco solutions, but you may not be aware of VMware and white box switch options. Each vendor brings significant benefits to you, but that doesn't mean it's a template solution. When a lot of architectures are put together, the situation is complicated.

Fruehe: What data center SDN lacks is timing. It's a bit like buying an electric car. When you think electric cars are great, you might consider using one the next time you go to buy a car. But if your car works, you won't throw it away immediately to buy an electric car because it's more efficient. SDN brings a lot of benefits, but it won't make consumers accept it all at once. This is the biggest challenge facing SDN, and it needs a mature time.

Original link:

https://searchnetworking.techtarget.com/feature/Reality-check-How-has-data-center-SDN-evolved