Amid daily rumors about the ups and down of the latest software-defined networking (SDN) startups, there is a raging debate going on about the health of the market and how many startups will survive.

I have the answer: Only a few of them will survive. This isn't anything illuminating; it's the way all startup markets work. Data shows that at least 75 percent of venture-backed startups fail (and no, WSJ, it isn't really a secret). Isn't this always the case? There will be violent shakeouts, and at the same time a few great companies will be created.

Nobody said it was easy.

It's important not to confuse trouble in startup land with the success of the SDN market in general. Let's use the Internet as an example: In the late 1990s, there was an obvious and spectacular Internet bubble. Ridiculous acquisitions were made, spectacular IPOs occurred, and then everything crashed.

In 2001, many concluded — wrongly — that the Internet was done, as things had gotten out of whack. There were bankruptcies and flameouts. Webvan and Pets.com no longer exist. But as it turns out, the Internet continued to grow and actually became more successful than previously imagined, spawning even more successful companies. Google, a mere baby in 2000, is now worth $336 billion. Facebook is worth $200 billion. And Amazon, originally viewed as a "gimmick" Internet company that some believed wouldn't survive the crash, is now worth $136 billion.

During that period, thousands of companies went out of business, but the entire market was successful.

So, there are important distinctions to be made between secular technology trends and the ability of startups to capitalize on them.

The secular trends in the network continue to be in place: Open standards, cloud computing, open source, and virtualization. These trends, in fact, appear to be gathering momentum.

It won't translate into riches for everybody that puts "open networking" in their business plan. There is no doubt that 2014 was a tough year for SDN startups, whether they were selling controllers or NFV applications — but it was certainly more difficult if you had originally set out to be an SDN controller company.

Earlier in the year, I penned a report on the SDN market called "SDN Revolution." In it, I revealed that more than $600 million has been pumped into the SDN startup market. I'm actually in the process of revising that to more than $700 million. The money is still flowing. Are the VCs dumb?

These numbers are frequently met with skeptical cackling. $600 million for free open source software! Bwaahahahha! But stop laughing, already. By my calculations, there have been more than $3 billion in SDN startup exits. So already, four times the amount of value has been created in exits as was put into the market.

So, you can stop your cackling now.

Virtualization of the networking world will continue, with many twists and turns, regardless of what happens to the latest hot-shot startup.

Earlier in the year, I gave some advice to SDN startups. One of them was to prepare for a long battle, because this market represents such a big change. And there is no doubt that Cisco and VMware have been resurgent in their attacks.

It's still going to be a long battle. It's going to take creativity, persistence, and marketing changes for each startup to find their niche. Each company is going to have to zero in on how to attack the market and pick off customers one by one.