Another aspect of network automation relates to network planning and optimization, and there is a move to go beyond SON (self-optimizing networks) and apply AI techniques to create a responsive network which orchestrates physical and virtual resources, and adds a high level of intelligence to the automation.
Mirko Voltolini, VP of technology and architecture at Colt’s technology services group, said automation and AI were important areas of development for the enterprise fiber operator. He told the SDN/NFV Congress in The Hague that Colt has created a new AI-driven conceptual platform called Sentio, and is aiming to develop fully automated service management capabilities for network functions such as traffic flow classification, fault prediction, path optimization, capacity management, security, intelligent bandwidth-on-demand, and many others.
Like many areas of virtualization, automation and networks-on-demand, this is being pioneered first on a fixed-line platform, partly because of the added complexity of managing moving objects, but the MNOs are watching such initiatives closely and planning similar approaches, often to span wireline and wireless networks. There are two foundations to Sentio – automated scaling of VNFs (virtual network functions) to launch, adapt and dial down services as required; and an analytics engine based on machine learning and AI, which would feed intelligence into Colt’s network controllers.
Colt told LightReading it has issued a request for information (RFI), and is hoping that there will be existing technologies, plus new ideas from responding vendors, to start the telco on its way. Vendors are starting to leap on the AI/network management/automation opportunity – Nokia has several elements in its self-driving telco platform, for instance, while IBM has been working with Accanto Systems in this area.
Unlike some operators, Colt is enthusiastic about the role of ETSI ISGs (Industry Specification Groups), mainly to help identify use cases, and to provide blueprints which can unify solutions. For instance, Voltolini said Colt had paid a price for being an early mover in NFV, in terms of an inefficient process to on-board new VNFs. It is addressing this by using ETSI NFV specifications to move to common orchestration, VNF manager and Virtual Infrastructure Manager (VIM) layers, and that will also underpin Sentio.
He also welcomed other ETSI activities in this area, including the new zero-touch group, but particularly the use case-oriented approach of another ISG, formed earlier this year and called Experiential Network Intelligence (ENI). It was largely the initiative of Ray Forbes of Huawei, and other founder members include Verizon, China Telecom, Samsung, Xilinx and two research instiitutes (the China Academy of Telecommunications Research and the Hong Kong University of Science and Technology).
ETSI ENI’s remit is to define an AI-based context-aware system based on the ‘observe-orient-decide-act’ control model. This year, it will develop use cases and requirements for telco network AI, which it will model in 2018. It will then move on to testing and automation in 2019-2020.
As in other areas of the new software network, ETSI’s work is either complemented or challenged – depending on your viewpoint – by open source activities, and often these are emanating from the Facebook-initiated TIP (Telecom Infra Project). TIP is to launch a new working group focused on AI and machine learning for telco network management and optimization, in November.
This will be co-chaired by Deutsche Telekom and Telefonica and according to Axel Clauberg, DT’s VP of aggregation, transport and IP, it will be “about how machine learning is applied to network management and how artificial intelligence can be used.” That ties into the overall TIP goal of reinventing the operator network around software, open source and commoditized hardware and introducing an entirely new cost base for telcos. Echoing that, Clauberg said that DT’s goal is to “get better at finding radical approaches to capital efficiency”.
On the ETSI side, Forbes said in a recent interview that many operators are incurring unnecessary cost by over-engineering their networks (adding surplus capacity to cope with peak loads) and running services at the least efficient times. He estimates that networks are typically 50% over-engineered, but that AI could reduce that to 10% by allocating resources more dynamically and efficiently according to workload, time of day and geographic patterns.
This would further enhance the resource efficiency impact which operators are already envisaging from virtualization and software-defined networking (SDN), which make it possible to dial network resources up and down on-demand (by calling up new instances of virtual network functions), while an SDN controller orchestrates those VNFs across the whole network.
AI would provide a new level of intelligence and automation for those SDN decisions, while also using machine learning to understand usage patterns in more detail, relating these to network resources, and predicting any faults and failures before they affect user experience. A
As China Telecom’s Haining Wang, vice chair of the ENI ISG, put it, while SDN and virtualization are helping to make networks more flexible, management complexity remains as high as ever – it is just transferred from hardware to software.
“It’s making resource management less human dependent in terms of knowledge,” Forbes said. “You want to have a business analyst who just performs data models; you don’t want to have a lot of pre-thinking and management analysts that analyze what to do.”
At the heart of ENI’s approach will be context awareness – the goal of many 5G-related projects, which envisage a whole new wave of mobile experiences, and revenue-generating services, which derive from the networks’ precise knowledge of a user’s context, preferences and habits.
The group says it will define an AI-enabled context aware system based on the ‘observe-orient-decide-act’ control model, which allows a system to adjust the services it is offering according to changes in user requirements, environmental conditions and business goals. This system will result in standardized models for using AI to manage a network and intelligently control service delivery and user experience.
The automated system will learn from previous experience and so add intelligence to automated network planning and monitoring, reducing opex, improving capacity usage, and boosting user experience, to a greater extent than just using automated technologies like self-optimizing network (SON).
The group will also study network telemetry, big data mechanisms, machine learning, and how to simplify and scale complex device configuration and monitoring.
“The unique added value of the ETSI ISG ENI approach is to define new metrics to quantify the operator’s experience; this enables the optimization and adjustment of the operator’s experience over time, taking advantage of machine learning and reasoning,” said Forbes when the ISG was announced in April.
Adding AI to the mix promises to make automation intelligent, and so deliver better results, but it will also slow down the process of getting to the self-driving, fully software-driven telco network, and some carriers are very impatient to reach that stage (see previous article). But AI and machine learning cannot be rushed – they are of their nature slow to perfect because the models need to learn until they reach the stage where they can make quicker, better decisions than a team of trained experts. “You need to design a model and spend a bit of time teaching the model how to do things,” Forbes said.
Forbes believes the dream of “fully AI-based autonomics” is far away, though ENI is taking steps in that direction. “The ultimate aim is to improve the user experience and simplicity to have intent-based network indications in line with customer demand for new revenue,” he concluded.
Full context awareness combined with automated, AI-enabled network management is at the heart of many 5G visions and operators like Telefonica, Verizon and others are starting to take steps towards that. However, it will not be an overnight process, and it will be essential, in every aspect of automation, that operators do not sacrifice openness and future-proofing, in their race to cut costs and support new service models.