x
NFV MANO

BT Wrestles With NFV Orchestration Confusion

UK fixed-line incumbent BT says there may be a need to go back to the drawing board when it comes to NFV orchestration because of complications that have arisen since standards body ETSI first outlined its vision of the NFV management and orchestration layer (MANO).

While a number of service providers have set out ambitious plans for the introduction of NFV technology, BT Group plc (NYSE: BT; London: BTA)'s Andy Reid, the operator's chief network services architect, argues there is a need for "more theory to solve practical problems." (See Vodafone Calls for End to Five Nines.)

There is some confusion about the roles that different elements of the NFV architecture are supposed to play and what Reid calls a "lack of precision" regarding the responsibilities of the various functional blocks.

"There is a lot of opportunity to work on theory," said Reid during a keynote presentation at Light Reading's OSS in the Era of SDN and NFV Conference in London on Thursday afternoon. "One of the projects we're involved with is about developing code to illustrate the role of orchestration in automation and superfluidity."

Asked to explain the concept of "superfluidity," Reid indicated it is actually the name of a European Union project looking at the future use of virtualization in radio access networks, among other things. "If you start including FGPAs [field programmable gate arrays] there are a whole set of capabilities that become more programmable and that creates a new optimization problem," says Reid.

Launched in July, the superfluidity project is receiving nearly €8 million ($8.7 million) in funding under the EU's Horizon 2020 research and innovation program and is set to run until December 2017.

While orchestration issues are not preventing BT from taking a number of practical steps in areas where there is no immediate need for a "full-blooded solution," Reid emphasizes that a lot of work needs to be done on developing standards in future.


For more NFV-related coverage and insights, check out our dedicated NFV content channel here on Light Reading.


Encouragingly, for Reid, ETSI is setting up an inter-SDO (standards development organization) on the subject of information models -- which describe infrastructure and VNF (virtual network functions) requirements -- and the BT executive believes the group could play a crucial role in this area, pulling together standards bodies such as the Open Networking Foundation and MEF .

Another EU project being managed under the Horizon 2020 program and also involving BT is Sonata, whose explicit task is to produce an open-source orchestrator, according to Reid.

Like the superfluidity scheme, Sonata was launched in July and will run until January 2018.

According to the EU's own website, the project will cost about €8.3 million ($9.3 million), with nearly €6.7 billion ($7.3 million) of that coming from the EU itself, and involves a number of other technology companies and research institutions, including Alcatel-Lucent (NYSE: ALU), NEC Europe and Telefónica .

Reid expressed regret that he did not push European Telecommunications Standards Institute (ETSI) to take a more holistic view of the NFV architecture and processes when the organization was first developing its blueprints.

"I wish I'd pushed harder for that," he told attendees at the OSS event. "Shouldn't we take an end-to-end process perspective and understand how they'll work logically and then map that back to the architecture?"

Instead, says Reid, the industry has ended up with more of an "API [application program interface] specification" that is not without merit but is clearly insufficient.

"We're not going to get the interoperability and the end-to-end systems we want," he says. "These processes are not going to join up just with what we've got."

— Iain Morris, Circle me on Google+ Follow me on TwitterVisit my LinkedIn profile, News Editor, Light Reading

axelc 11/11/2015 | 3:28:30 PM
Implementation experience We should all keep in mind that writing standards requires implementation experience. Rough consensus and running code worked out for the IETF, and are definitely required here, too.

Look at the Hyperscale operators, they are not really waiting for complex standards to be developed. They implement, learn from it and go into the next iteration. TelCos need to follow the same paradigm.

 
NeilMcRae 11/9/2015 | 6:19:00 AM
SDO Please no more standards forums!!!
HOME
Sign In
SEARCH
CLOSE
MORE
CLOSE