Telecom IQ
Showing results for 
Search instead for 
Did you mean: 

Five Essentials for NFV Success


As we work towards the end of 2015, many NFV deployments are progressing from the Proof of Concept (PoC) stage into actual commercial launches. NFV is one of the most quickly-adopted technologies ever seen in telecom and even in tech in general. It is racing its way through the Garner hype cycle, and "crossing the chasm" would better be described as "leaping the chasm" when it comes to NFV.


So, as CSPs and their vendors begin to launch commercial NFV, it's perhaps a good time to check the best practices, and discuss the Five Essential Principles of NFV success.


NFV Executive Leadership Forum

Every quarter, the Telecom Council of Silicon Valley (a regional club with over 130 members including the world’s largest Communication Service Providers, CSPs) collaborates with us to organize a high-level executive roundtable discussion on NFV. Attendees represent leading NFV minds from CSPs, telecom infrastructure vendors, IT companies, virtualization companies, off-the-shelf server companies, and start-ups. The participants share and learn about the state of NFV, the roadmaps, and the challenges ahead. In the recent August meeting, Jeff Edlund CTO of Communications, Media and Entertainment for HP, wrapped up the discussion with his summary of the Five Principles.


Jeff explains that in his extensive work inside of the NFV standards bodies, PoC trials, conferences, and observations, he has whittled down what the successful CSPs do, and what the unsuccessful ones are not doing. The following summarizes Edlund's spoken comments:


1. IT and Network Ops Teams Must Come Together 

Consistent with the lessons of virtualization in the enterprise IT space, if CSPs want to succeed with NFV, they will need to bring their Network and IT units together in collaboration on the project. In fact, these two groups, long term, may even merge. In the short term, a shared infrastructure team with representation from both sides should work on NFV PoC or early deployments.


2. CSPs Must Leverage the Insight of Product Marketing 

Product Marketing and Network groups never exactly see eye-to-eye, and in fact people who deeply understand technology may not respect the views of people who simply understand consumers. But those differences must be cast aside. Product Marketing needs to be an active partner in NFV PoCs, because it is fatal to build tech "just for the sake of tech". With Product Marketing, initial PoCs can be selected and built to offer solutions for which the end customer will pay. It is in proving commercial viability that NFV will prove its worth – and that requires cooperation from marketing and engineering. This requirement is consistent with what IT is learning in the field of DevOps.


3. The Commit Level Must Go Above the CIO and CTO 

Fairly simple point to grasp here. Powerful interests within the company must be aligned if success is to be the result. NFV, long-term, will result in major structural and functional changes within a CSP. This carries career-level existential risk to many staff and senior executives. Therefore, the entire C-level suite and the board must be on-board for NFV to ensure every hand at the helm of the ship is steering towards the same course.


4. CSPs That Contribute, Succeed

Like learning a foreign language, it's not enough to watch people speak French to learn it, otherwise every college kid with a Eurailpass would come home citing Voltaire in original prose. Nope, you have to get your feet wet, you have to immerse yourself, and you have to actually practice speaking to learn how to do it. Edlund says the same of NFV: "CSPs that 'monitor' or 'participate' in the standards discussions will fail, and those that contribute will succeed. Like French, you just can't really learn the lessons by monitoring a discussion.

Which CSPs are actively participating, contributing, and learning right now? This list of On-Going PoCs from ETSI gives a good idea of the leaders.


5. Open Source 

NFV is about not getting locked-in to vendors. This means that VNFs, platforms, orchestration, etc. need to interoperate in a multi-vendor environment. Therefore, NFV demands a high level of openness than other technologies. The companies that are contributing to, and leveraging the Open Source solutions will have long-term advantages in NFV.



It's important to note that Edlund's argument was not just that CSPs will have better outcomes if they adhere to the Five Principles above, but that the inverse is also true – companies that don't do the above will have worse odds with each Principle they violate. Edlund argues that you aren't picking from the five like it's a menu – you have to do them all!


Other Highlights of the August Meeting

There were other notable discussions in our two-hour meeting. Observing Chatham House Rule, we'll just present the thoughts, not the name of the speaker:


  • Historically, CSPs have had a "Department of NO!!". This normally would be the Network Planning or Network operations department. This group, responsible for uptime as opposed to innovation, would routinely say "No" to new service ideas from Marketing or other groups. But NFV is all about empowering the Dept. of No! to get to "Yes". Startups may have encountered this Department, and have certainly felt the "no", but the revelation was that the Dept. is just as tough on internal innovation as external. Read the Telecom Council Blog post about the Dept. of No.
  • Our August group discussed Business Models around NFV, which included models for new service offerings like MVNO, subscriber data pricing models like tablets that are sold with variable data plans, and Microservices delivered almost like from an app store: a remote DVR, HD Voice, conference bridges, etc.
  • The discussion veered into a tangent of the risks of just "virtualizing the legacy hardware structure". When radio was invented, the first thing they did was read newspapers on the radio. That model is wrong – don't just take the legacy way of doing things when you have a revolutionary platform. It's time to re-think the architecture from the ground up. There will certainly be inspiration and even wholesale components taken from legacy, but it needs a fresh evaluation.
  • The last major topic we fit into the meeting was the issue of Security, and whether NFV, overall, increased security risks for the CSP, or reduced them. The general conclusion is that risks will exist, but NFV offers the ability to have security VNFs. Heuristics and analytics can be used to parse for unusual activity, and the ability to update, change, or enhance security is a virtue of NFV. Virtualization also reduces the risks of attach, and virtual machines can be shut down, and spun back up as clean versions quickly.


Conclusion and Invitation

The summary above represents just a portion of our August NFV Forum. Things are moving fast in NFV. With these quarterly meetings, we are tracking the shifting landscape of NFV. We'll continue to report back some of our findings. For every meeting, we try to keep the head count at 20, and try to bring in fresh people and their ideas. HP is committed to "Open", so the meeting attendees are invited and administered by the Telecom Council, and open to competitors of HP. If you are interested in attending an upcoming quarterly meeting, contact the Telecom Council of Silicon Valley.




0 Kudos
About the Author


Jan 30-31, 2018
Expert Days - 2018
Visit this forum and get the schedules for online HPE Expert Days where you can talk to HPE product experts, R&D and support team members and get answ...
Read more
See posts for dates
HPE Webinars - 2018
Find out about this year's live broadcasts and on-demand webinars.
Read more
View all