China Mobile and ARM combine on network function virtualisation

China Mobile and ARM will collaborate on network function virtualisation (NFV) thanks to a deal signed with processor provider Cavium and network software platform supplier Enea in an effort to drive larger-scale deployment of virtualised networks.

Under the deal, the companies will make use of China Mobile's open NFV test lab, Enea's NFV Core platform, Cavium's ThunderX datacenter server processor, and ARM's 64-bit processor.

This platform, hosted as part of China Mobile's telecom integrated cloud, will then trial several NFV test cases, including virtualised customer premises equipment (vCPE), virtualised broadband remote access server (vBRAS), virtualised evolved packet core (vEPC), and virtualised IMS (vIMS) network functions.

ARM said it is working towards enabling open NFV so as to provide cost-efficient computing power for data networks globally.

"The network pipeline must be expanded, at scale, to support new computing across all markets in the most efficient way possible," ARM VP and GM of Business Segments Group Networking Servers Noel Hurley said.

"Our ecosystem delivers this efficiency through well-integrated solutions that demonstrate the performance-per-watt, density, and TCO advantages provided by ARM technology."

Enea's open source based NFV (OPNFV) Core is built on OpenStack, able to deploy and manage vCPE network functions in both offices and datacentres. According to Enea, its OPNFV platform is the first virtualised platform able to use ARM servers out of the box.

China Mobile had chosen to implement NFV with OpenStack at the end of last year, alongside fellow telecommunications providers AT&T, Deutsche Telekom, NTT Group, SK Telecom, and Verizon. OpenStack allows telcos to run network appliances as virtual appliances on commercial off-the-shelf servers (COTS) via the cloud.

"We're seeing incredible progress and rapid adoption of OpenStack by the telecom industry, and, in return, the contributions the telecom industry is making to OpenStack have far-reaching value for other businesses with challenging requirements such as high-availability, scaling, fault management, and a diverse footprint of facility sizes and locations," OpenStack Foundation executive director Jonathan Bryce said in October.

This came after China Mobile signed a one-year €1.36 billion agreement with Nokia in 2016 for cloud services provision, as well as mobile, fixed, IP routing, optical transport, and customer experience management technology.

In addition to working on NFV, ARM has also been pushing into the automotive market, last month unveiling Mali-C71, an image signal processor that processes raw pixel data to be displayed to a driver or processed by computer vision algorithms.

Mali-C71, the first product to come out of ARM's 2016 acquisition of Apical, enables ARM's partners to process images from every camera on a car within one system on a chip (SoC), instead of one SoC per camera.

Mid-range cars are expected to have three cameras built in by 2023, while luxury vehicles are forecast to have 10 cameras.

"The reason people add cameras to cars is to make the car safer, either to make it capable of autonomously driving, to make the car aware of its surroundings, or to simply better enable visibility for the driver," said Richard York, senior director of ARM's embedded segment.

This followed ARM in March unveiling DynamIQ, a multi-core micro-architecture that will serve as the foundation for its new Cortex-A processors, allowing it to support intelligent systems from the cloud to the network, edge devices, and automotive technology.