Grift’s like anything else, Roy. You don’t stand still. You either go
up or down. Usually down, sooner or later.
Lilly Dillon from “The Grifters”
At Interop this month, every vendor had product sheets that claimed, “Now with SDN!” It’s the latest industry buzzword and I started to recall some previous one-hit wonders from the past. Remember IPAM? How about SEIMs? They were going to save our organizations, optimize the efficiency of our staff and where are they now? Don’t get me wrong, SDN or Software Defined Networking sounds great, just like cold fusion (the nuclear reaction, not the software) and flying cars.
In a discussion with Ivan Pepelnjak, I told him my biggest problem is that it’s reminiscent of the wireless LAN controller industry. They’re based on IETF standards with RFCs and everything! Everyone is supporting CAPWAP, there’s even an open source project. So WHY can’t I have full vendor interoperability with my WLAN equipment? While I agree with his assertion that SDN and WLAN controllers differ in implementation, they are both similar in the abstraction layer they present to the user.
As I listened to vendor pitches for their respective SDN implementations, I realized this was just one more in a series of industry “long cons.” For those unfamiliar with the world of con artistry, according to Wikipedia:
A long or big con is a scam that unfolds over several days or weeks and involves a team of swindlers, as well as props, sets, extras, costumes, and scripted lines. It aims to rob the victim of thousands of dollars, often by getting him or her to empty out banking accounts and borrow from family members.
Sound familiar? SDN, the ultimate grift. Better than anything out there right now and the marks have cash to burn. It didn’t have to be this way, but it seems that’s just what happens with great ideas in networking anymore. Ultimately, each vendor’s proprietary implementation will probably wind up on the dung heap along with all the marketing propaganda, but only after our organizations have dumped heaping piles of cash into products promising us a brave new world of optimized and easily configured networks. In all likelihood, network engineers will keep struggling with automation, orchestration and centralization, feeling swindled, but hoping for the next big thing to solve their problems.