Wingspan

Migrating to SDN

by John.McHugh on ‎10-01-2012 09:25 AM (4,460 Views)

The industry is all abuzz about Software-Defined Networking (SDN). As with every new technology trend that is unproven, unimplemented, and largely theoretical, SDN seems to solve essentially any problem a network architect and network operator might have.

Pundits claim it will commoditize networking equipment and save customers millions of dollars. It will squeeze all of the unused bandwidth out of existing networking infrastructure and deliver lightning-fast performance. It will allow the network to be completely programmed for optimum application performance, regardless of what application is running and what other applications might coexist on the network. In fact, the only thing greater today than the promise of SDN are the promises of those who are pushing SDN.

One of the critical survival techniques in this industry is knowing the difference between flash-in-the-pan hype and technology with real merit that has unfortunately been sucked up into the hype tornado. The fact of the matter is, this industry hypes everything. However, that hype almost never correlates to what is good for customers and what is a dead end (ATM to the desktop and open source networking stacks, to name just a few).

SDN is clearly a concept that has merit, but the actual implementations are well behind the hype cycle. Luckily for customers, there is an alternative that allows them to position themselves for the future without having to jump headlong into unknown waters.

At the annual Brocade Analyst and Technology Day, on September 12, we announced that we would begin shipping software for the Brocade MLX Series routers that will enable customers to gracefully prepare for the days of actual SDN deployment without having to rip and replace their existing network infrastructure or settle for a lesser product today.

Brocade’s true Hybrid Mode capability enables customers to utilize the same router for both traditional and software-defined networks. Customers can maintain their traditional forwarding policies—along with all the established operational processes to keep that traffic flowing—and then selectively enable OpenFlow on particular data flows. Unlike some other vendors’ offerings that force an “either/or” answer, Hybrid Mode lets customers introduce OpenFlow into their environment at their own pace.

The concept of building a completely SDN-driven production environment today is still a bit of a stretch. The reality for the next several years is that customers will operate networks using traditional network design and programming practices, but will begin moving to selective routing of specific flows as the technology matures. Ultimately this may lead to widespread deployment of solutions that can operate solely in an SDN mode, but that transition will likely take years.

Brocade has always designed our networking solutions for the most demanding environments, and that innovation gives our hardware enough headroom to support Hybrid Mode without performance degradation. As a result, our customers have the best network for today and the best network for tomorrow—without suffering through a forklift upgrade and unnecessary risks and worries.

Make no mistake: SDN is an extremely promising technology. As a pure-play networking company, Brocade has been actively supporting and investing in SDN for years. And as a company that defines ourselves based on our ability to provide customers with the most agile and efficient solutions, we will continue to ensure that the Brocade products customers buy today are still relevant as this new technology matures.

Comments
by
on ‎10-03-2012 12:32 AM

I might not have too much knowledge regarding SDN, but I totally agree with your article. Too many "evangelists" see SDN as the  magic wand to solve "all network related issues".

by
on ‎12-14-2012 09:55 PM

SDN didn't even have a name last year at this exact time; literally.    

But yes people were just coming up with its concept at that time; primarily Stanford.