Artificial intelligence is key to self-driving networks

Jason McGee-Abe
Published on:

The body that led the way to virtualisation has set up a group of industry experts to create standards for how artificial intelligence (AI) can help carriers improve network services and security.

The European Telecommunications Standards Institute (ETSI), which wrote the original specification for network functions virtualisation (NFV), has set up its experiential network intelligence industry specification group to handle the task.

Kireeti Kompella - Juniper Networks
Kompella: Self-driving networks will become absolutely mandatory
The key is to differentiate automation and autonomy to fully comprehend the task ahead, Kireeti Kompella, CTO of Juniper Networks’ development and innovation team, told Global Telecoms Business.

He sees many parallels between the evolution of cars and where networks can get to. "Years ago, cars were painfully manual, but we've made it much more convenient and 14 years ago we started looking at self-driving cars, which is an absolute disruptive change."

In order for self-driving networks to become a reality, vendors and providers must fully collaborate, he says. Interestingly, ETSI’s AI initiative, which has an initial two-year work programme, is coming from vendors, ETSI director general Luis Jorge Romero told GTB at Mobile World Congress. Huawei, Samsung and Xilinx are the initial members, along with universities in China and Luxembourg. But that’s likely to grow, as did the membership of the NFV group after it was set up.

Romero said: “This group of companies came to us, saying it’s about time that we could use AI to help us provide better help to operators with network resources. We will be taking into account the impact of 5G as well.”

Kompella added: “One place that we really see a need for self-driving automation is 5G and even more so the internet of things (IoT). Meaning the sheer scale of things, trying to corral things with humans operating is just not going to be effective, and the amount of data that will be coming back.” He added: “I’m a huge fan of automation but I want to create the type of disruptive change that occurred in the automotive industry to managing networks. Self-driving networks is not about incremental additions of automation but it’s about a disruptive change. It also has an economic benefit, as the biggest cost today of managing networks are the operations.”

Kompella states that there are five technologies that are needed for self-driving networks to succeed. “It starts with the real-time, highly optimised telemetry,” he said. “Secondly there will be multi-dimensional views to get deep into the network, correlating all the data to attain a better understanding of what is happening.”

LJ Romero Quote GTB mag march-april

How we interact with the network is another component, he said. “Today we have a tendency to go to the bots and type command line interfaces and tell the bots how they should behave. We should be able to get to a place when we tell the network what we want it to do and let the network figure out how to do it. Declarative intent is needed too and the latest technological key is machine learning.”

Juniper Networks has acquired a company called AppFormix, which will bolster its cloud portfolio with machine learning and telemetry technology. “I’m looking at how we can apply machine learning for server management towards network management.”

The telecommunications industry is investing heavily in AI. A few examples include Deutsche Telekom forming new AI and security partnerships to bolster its offering to businesses.

“Systems with artificial intelligence provide ideal impetus for the digital transformation process, and they are playing an increasingly important role in business competitiveness,” said Reinhard Clemens, CEO of T-Systems.

Last year, MIT built an AI-based cyber security platform, called AI2, that security researchers say can detect 85% of cyberattacks.

Artificial intelligence represents a fundamental technological shift that opens up for new opportunities, efficiency and increased competitiveness. However, when someone attacks a network, that’s when an immediate response is needed and it’s urgent. Can we mitigate threats quickly, particularly if people are trying to extract data?

“Combine that with IoT, 5G and the scale that the networks are going to get to, I think that’s where self-driving networks will become absolutely mandatory,” said Kompella. GTB