 | Issue: | North America II 2015 |
| Article no.: | 8 | |
| Topic: | Keeping up with the Kardashians: Virtualized Network Functions and the ICT sea-change | |
| Author: | Franklin Flint | |
| Title: | CTO | |
| Organisation: | Telecommunications Industry Association (TIA) | |
| PDF size: | 196KB |
About author
Franklin Flint joined Telecommunications Industry Association (TIA) in March 2015 as Chief Technology Officer, leading the technology, standards, and numbering activities in the association. He also creates thought leadership, technical strategy and industry insights.
Franklin previously helped develop the telecommunications vertical strategy at Dell Inc., and has extensive experience in Enterprise, Data Center, Networking, Storage, and Server technologies after spending 20 years in the commercial sector working with the world’s largest enterprises to develop ICT strategies and solutions. He has had dozens of thought leadership pieces published, developed standards based solutions for multiple vertical markets, and created an extensive collection of speaking pieces focused on the future of the technology market.
With an extensive history in developing global alliances, joint initiatives, ecosystems, and industry visions which improved the overall ICT community, Franklin delivers his industrious and unique skill set to the TIA.
Article abstract
The promises of NFV and SDN are extremely compelling, but there are many issues to consider as well. With the adoption of virtualized networks, the economy for solutions will be open to anyone with a great idea. Instead of going to a traditional vendor for network services, service providers could go to a startup offering a new way to solve a problem and often a lower entry price. The shiny new ideas can be blinding, however, much like the app store for your smartphone more useless apps than high value apps will be offered. There is always risk in going with a new company’s product instead of those offered by a tested and trusted partner.
Full Article
There is a sea-change afoot in the telecommunications industry and it is hard not to notice. It is big and it’s being driven by data. According to TIA’s annual Market Review and Forecast, wireless subscribers spent more on data than on voice for the first time last year.
Speaking to attendees at TIA’s 2015 Network of the Future Conference, Andre Fuetsch, senior vice president of architecture and design for AT&T, said it best: “If you look at this demand curve now, voice traffic is a trickle. Now the traffic capacity engineer has to worry about the latest Kardashian streaming video, who has been watching House of Cards. And the biggest event we worry about is when Apple releases their next big iOS upgrade because you never know who is going to push the button to download now.”
It is necessary for service providers to evolve to meet users demand for data, especially in light of the ongoing growth of video streaming. This will require providers to rethink how they build their networks.
Building Networks from the Bottom Up is Out
In the old business model, telecom service providers would offer most services by purchasing a device, such as an IP multimedia subsystem appliance or evolved packet core platform, which had the service capabilities already installed by the manufacturer. They would then hire brilliant engineers to find the best way to integrate the new hardware into the existing network and determine ways to reduce or eliminate any downtime during that process. It was standard for a new service rollout to take up to two years from pulling the trigger to being fully deployed and offered to customers.
Today, there are relatively well-tested Software Defined Networking (SDN) technology that can reconfigure the network on the fly without physically altering connections and physical interfaces. The ICT industry is also experiencing a massive revolution with emerging technologies such as Network Functions Virtualization (NFV). NFV establishes communications networks from the top down by moving services from hardware firmware to software running in network cloud environments.
In the not so distant future, the infrastructure of a network will be chosen to ensure connectivity and appropriate amounts of processing and bandwidth instead of being driven by the schematics and hardware purchases for particular services. Telcos will build data centers – large and small – filled with industry-standardized servers, networking, and storage, which will be distributed throughout their networks.
The many servers and switches will be run by cloud software platforms that enable network functions to efficiently operate on the servers. The software will also allow network controllers to create the virtual pathways necessary to chain numerous network functions together in the right order and at the right performance and uptime requirements. Service providers will continuously evaluate their networks and purchase virtual network functions in the form of licensed software to deploy in the data centers, effectively creating a communications network.
Once fully realized, the capabilities of NFV and SDN will change the way communications companies offer services to consumers by:
1) Breaking apart the supply chain for building networks with the hardware being evaluated and purchased, in most cases, separately from services;
2) Increasing the capacity of a network without choosing a vendor for each service being offered;
3) Speeding time to market and propelling innovative new services being trialed and scaled-up if successful, or quickly removed if not, in weeks or months rather than years; and
4) Requiring less intensive and expensive engineering to design and service networks, deploy capabilities or increase capacity for services.
Beware the Carpet Baggers, but Welcome the Innovators
The promises of NFV and SDN are extremely compelling, but there are many issues to consider as well. With the adoption of virtualized networks, the economy for solutions will be open to anyone with a great idea. Instead of going to a traditional vendor for network services, service providers could go to a startup offering a new way to solve a problem and often a lower entry price. The shiny new ideas can be blinding, however, much like the app store for your smartphone more useless apps than high value apps will be offered. There is always risk in going with a new company’s product instead of those offered by a tested and trusted partner.
But there is also risk for service providers in maintaining the course with the same suppliers they’ve always used. Innovation often comes from competition and to drive competition new suppliers need to be at least somewhat successful. The innovators of the next generation of network solutions are more likely to be those who are new to the market, which may include startups or IT companies fresh to the telecom industry. There have been significant solutions developed in the IT sector that are not currently used in the telecommunications arena but may be applicable to new networks based on NFV and SDN.
In many ways, the capabilities and flexibility of NFV will assist service providers in accepting this risk. With a virtualized communications network, new services can be tested on a real network using real traffic before being fully deployed or even purchased. Within weeks, it will be possible to pressure test a seemingly great idea to see if it passes, fails, or needs improvement. In addition, excess capacity in service providers’ data centers could be given to researchers and academics to test ideas and experiment with new concepts, spurring further innovations.
NFV and the Bottom Line
There are exceptional opportunities to be pursued using NFV and SDN technologies, but not every service provider wants to take risk or be seen as innovative. Some are very comfortable with being safe and consistent in offering services to their customers at reasonable prices and with no gimmicks. NFV and SDN offer benefits to these companies as well.
Virtualization will change the model for purchasing, deploying, and servicing communications networks. With a virtualized data center approach to the network, services will be deployed and managed almost entirely from a central network operations center. Fewer trips to the field by experts will reduce the cost of operating the network. Field-based service work will be simpler and demand a less expert workforce to perform. To total, this will translate to a lower operational expense model.
Once a network is virtualized, it will be running on widely available commercial off-the-shelf (COTS) hardware. While COTS may have fewer assurances of uptime and hardware failures may occur from time-to-time, visualized network services will be self-healing, so customers will not experience downtime or reduced performance when a hard drive, power supply, or entire server failures. These components can be replaced or repaired by IT staff – rather than an engineer – or the relatively affordable support team a manufacturer may employ. The cost of operating networks will improve by leveraging the well-established and competitive support models developed over decades in the traditional IT world.
Ride the Wave
There are still some kinks to work out and trials to be completed before large scale NFV adoption. However, some of the largest players in the industry have started advancing technology and deploying NFV solutions in their networks, creating a financial advantage for suppliers developing NFV solutions.
It will mean a radical shift in the traditional ICT business model, but the benefits of improved communications networks and currently unimagined innovations via NFV will surely be realized by service providers, their suppliers and manufacturers, and the businesses and consumers they serve.
