I recently spent a week in Berlin attending the OPNFV Summit, the conference for the group building a common, open source platform for hosting virtualized network functions (VNFs) like firewalls, monitoring and load balancers. OPNFV’s goal is to deliver a standardized, open source platform for VNFs that carriers like AT&T, China Mobile, Deutsche Telekom, Telefonica and Vodafone can use to automate their services. The changes that OPNFV brings will ripple out to your business soon.
(Source: John Fruehe)
The Open Platform for NFV project is driven heavily by telecom providers. Yes, the phone company is behind improving your enterprise network. Some may think of “The Phone Company” as a landline monopoly…or the villain in the 1967 film The President’s Analyst. But in reality, telecoms aren’t stodgy old companies anymore. They are transforming themselves. No longer mired in the old analog and POTS services of the past, telecoms are now actually running the innovative public clouds that power today’s enterprise businesses.
Spending all that time with carriers and their equipment providers yielded some key observations.
- The Carrier Business is Changing: Carriers are seeing their businesses fundamentally change, and instead of sticking their head in the sand they are embracing the change. Yesterday was dial up, followed by cellular and mobile internet. 5G and IoT are the next 2 waves to impact them, and both will demand NFV, which is why they are driving this so hard.
- OPNFV is a Team Sport: Open source is, by nature, a collaboration with many responsible for the implementation of technology and features. Through the presentations, it was obvious that competitors are working together, not against each other, because everyone benefits from pervasive NFV platforms (and splintering is the death of these projects). This was an interesting contrast, as Brexit was occurring that week. The EU seemed to be splitting apart, and at the same time the telecom world was coming together.
- Open Source is King: Everyone kept stressing open source as the key to success. There was a strong push on identifying the contributors and recognizing the companies that were driving contributions. These contributions not only impact OPNFV but also upstream projects like OpenStack. “OPNFV is more than a reference platform, it’s a lighthouse or guide for the upstream people like OpenStack,” was one of the more telling comments of the event.
- Big Bets are Being Placed: I had a chance to sit down with a few of the vendors. As a hardware vendor, Huawei is in a position to be significantly impacted by network virtualization. But instead of viewing it as a threat, they are exploiting the opportunity, working hard to enable NFV. But it is not all clear sailing, as Huawei will still need to overcome some of their geographic market challenges. Red Hat was demonstrating some previous capabilities that are now starting to attract interest, because market needs and capabilities are finally catching up. With their past record, Red Hat is in a good position, but the devil is always in the details when it comes to building a profitable business on open source code. The landscape continues to change, with Lenovo becoming the newest member right as the meeting was starting; project momentum also brings more competition, which is a good thing.
- Orchestration Will Be the Next Battleground: Two different initiatives OSM and Open-O are trying to create an orchestration layer for NFV. What this points to is the clear need for more standardization on how all of these VNFs get deployed. While multiple ways to solve a problem are great (does your company use just one database?), there are compelling reasons to bring these initiatives together and standardize some of the interfaces. Many of the NFV vendors are smaller startups who may not be capitalized enough to develop for two different orchestration layers. Driving some commonality here will be important in helping not only the vendors developing the solutions but also the end customers who will eventually deploy the solutions.
OPNFV has only been in existence for two years now and they are impacting the market. Because most of their direction is coming from customers (carriers, telecoms, cloud service providers) instead of vendors, there is an immediate buy-in for the work that is coming out of the organization. As the project matures and continues to churn out deliverables, it is imperative that they maintain their strict focus and not try to pursue every shiny object. Collaborative projects tend to see early momentum that eventually starts to slow if participants get distracted or fail to see the impact that their work is having.
So far, indications of the organization’s work are looking positive, and we urge them to maintain the pace of innovation. As an upstream organization they are in a good position to help negotiate an OSM / Open-O détente and help those two groups find some common ground and share some of the constructs that can make everyone more successful. If you had asked me ten years ago who would be driving IT innovation in the future, I probably would not have said that it would be telecom-led. Clearly the changes in their business models are forcing them to take more of a leadership role in deciding things. The future: brought to you by The Phone Company. Who would have guessed it?