Virtualization has been an up and coming platform for years now. And not just in the wireless space. And, depending upon which industry one looks at, the state of virtualization varies.
In the wireless space, virtualization promises a number of benefits, across the board. From efficient, real-time spectrum utilization to substantial savings in both equipment and manpower. And once virtualization becomes a standardized platform, equipment and software becomes widely available and economical, and deployments become commonplace, it should prove to be a huge enabler of 5G. But, we are not there yet. In fact, we are just at the starting gate.
At Mobile World Congress early this year, there was a lot of discussion about virtualization. And, as is the case with such events, companies took the opportunity to champion the technology and present their cases.
One example is that of joint project planned by Huawei, the GSMA, and China Mobile to develop an assessment framework for achieving carrier-grade network functions virtualization (NFV) for the wireless segment. It is championed as an effort to bring together a brain trust for developing a proactive, flexible and analytical network maintenance topology for telecom operators.
This is what I mean when I say we are just now at the starting gate. The industry has a good understanding of the concept but there is much to be overcome in both implementation and the challenges of just getting a technology out there that is implementable, workable and interoperable. So far, not much of this is real, as is seen through the current effort of just getting a standard maintenance platform hammered out. There have been several open source initiatives delivered such as the European Telecommunications Standards Institute (ETSI) open-source NFV framework for Management and Orchestration (MANO). This has been accepted by some Tier 1 communications service providers and they are actively participating in open source initiatives.
But on the downside, the industry has become fragmented with multiple open source initiatives, each claiming their own merits. These include: OPEN-O, AT&T‘s ECOMP platform, ETSI OSM, OPNFV, and others. While the Linux Foundation has combined OPEN-O and ECOMP into ONAP, it remains to be seen when the integration will be successful because the industry will still be left with a battle between OSM and ONAP. And, new initiatives, such as MEF LSO, are jockeying for position with their deeper integration of operations support system (OSS)/business support systems (BSS). All of this throws a roadblock out there.
And, even once this is ironed out, however it goes, there are still no real case studies other than trials and limited beta tests.
So, there are still looming questions. For example, can NFV/SDN virtualized networks meet and scale to the rapidly increasing network bandwidth demands using commercial off the shelf hardware in a data center? That is crucial since it isn’t realistic to have to change out all of the existing data center hardware and software.
And, will there be a single or multiple open source initiatives. If there are multiple, will they be compatible? Interoperable? Will there be reliable test platforms to verify and quantify actual performance metrics against specifications? For example, the promised latency and bandwidth optimization.
As well, legacy systems aren’t going away any time soon. So, any NFV solutions must be able to integrate legacy systems. And what about integration of new platforms and technologies, such as mmWave, HetNets and the evolution of technologies like Wi-Fi and LTE-U?
There is more, but one gets the picture that there are multiple issues that will have to be addressed and resolved before NFV get real traction. And, if the past is any predictor of the future, that can be a slow and painful process.
Nevertheless, NFV and its cousin, software defined networking (SDN) are the future of agile, ubiquitous wireless networks that will be made up of a lot of different technologies and platforms. So it is not a case of if but when NFV becomes tangible. And not likely very soon.