Lessons From the All-IP Transition
The telecom industry is so consumed right now by the transition to virtualized networks that it's easy to forget that another once all-consuming technology transition is nearing its conclusion.
Notwithstanding the FCC's all-IP trials with AT&T Inc. (NYSE: T), the transition to all-IP networks is very close (in telecom industry timescales, at least) to being finished.
How close? Close enough that some vendors believe they won't be selling any POTS equipment to telcos in another five or six years, according to the recent Heavy Reading Insider Report, " The Future for Telecom Network Equipment Providers ." (See The New Resonant Relationship Between NEPs & Their Customers and AT&T's All-IP Tests Won't Answer Key Questions .)
The report treats the all-IP transition as something that still affects the world of the network equipment providers (NEPs), while adopting the position that there is a high degree of uncertainty about the impact and future of that transition.
Basically, we now know how, and almost when, this story will end. That's strange to think about, given how long this transition has been ongoing, and how much some vendors dragged their feet along the way.
In their defence, though, some of the vendors that didn't get with the all-IP program from the outset were in many cases simply matching the slow pace set by service provider customers that wanted to get the most out of their legacy equipment and weren't feeling enough competitive pressure to move faster.
The transition to all-IP and the introduction of SDN and NFV are sometimes mentioned in the same breath, usually in describing SDN/NFV as the biggest technology evolution facing the telecom sector since the all-IP evolution began. (See Defining SDN & NFV .)
Beyond that, many people assume the SDN/NFV story will play out much more quickly than the all-IP transition has. There is every reason that it should: Competition at the service provider level is much more intense; carriers realize the value of dynamically responding to bandwidth demand; they can foresee internal operational benefits from virtualizing their systems.
But perhaps it would be wise not to let the all-IP transition conclude quietly. One of the biggest lessons NEPs should have learned from the shift towards all-IP networks is that they need to embrace the future even while supporting the network technologies of the past. They need to be flexible enough to support the creation of virtualized network environments even while maintaining "legacy" infrastructure in the field. They can't choose to put off virtualization just because they're still making hardware sales.
The transition to all-IP has been a painful one at times, and NEPs that didn't manage it effectively have paid dearly. (See Tellabs' Last Hope: Marlin's Optical Ambition .)
The next big networking transition doesn't have to be so painful.
