“The challenges of the telecom providers are enormous. The key is AI” – Techtime

December 4, 2022

Manfred Oficios, senior vice president of Juniper Networks worldwide, in an interview with Techtime: “The communication providers have less money and personnel, and on the other hand, they have to support more critical applications. We offer an approach of ‘autonomous communication networks’, like in the automotive world”

[בתמונה: מימין, מנפרד אופיסיוס, סגן נשיא עולמי בתחום ספקיות שירותים בג’וניפר, משמאל, איציק מלכה, מנהל פעילות ג’וניפר ישראל]

The gradual development from the fourth generation to the fifth generation is not limited only to increasing the bandwidth and speeding up the upload and download rates on our smartphone. This is a tectonic change in the architecture of the network, which is manifested in the transition to more distributed networks, which are designed to support not only human consumers, but also the countless IoT accessories deployed everywhere and critical applications such as smart cities and autonomous driving, which require zero latency and high reliability.

Those who stand at the forefront of this revolution are actually telecom companies of the older variety, the internet and cellular companies (ISP) that believe both in the deployment of the physical infrastructure and in the ongoing management of these demanding networks. Now, the communication providers are required to adapt to the new technology and models. One of the world’s leading companies in the field of hardware and software solutions for the telecom world is Juniper Networks. The Sunnyville, California-based company provides network equipment such as routers and switches, as well as software solutions for the management and security of software-based networks (SD-WAN).

According to Manfred Opificius, VP of service providers at Juniper, who spoke with Techtime during his visit to Israel last month, the communication providers face a series of complex challenges, both technological, economic and business.

“On the one hand, the communication providers have less skilled personnel and less financial resources, and on the other hand, they are required to support an exponential proliferation of applications and meet the demands of the customers for a high level of service. Among communication providers in the world there is a problem of personnel. The young hi-techists prefer to work in cloud companies or start-ups, rather than the traditional telecom companies. To this must be added the need to reduce energy consumption, both for reasons of cost and for environmental reasons.”

AI-based autonomous networks

According to Manfred, the key to dealing with these many challenges is automation. “These changes sharpen the need for automation. We offer a new approach of ‘autonomous networks’, like in the automotive world. That is, a communication network that manages itself completely automatically, and requires human intervention only in extreme cases, like in a Tesla car. This reduces costs and energy consumption, and removes from the human workforce the need to perform all kinds of tedious tasks that are part of the day-to-day management of a communication network.”

This transformation is performed by Juniper using artificial intelligence. This is where one of the most significant purchases made by Juniper since its founding, of Mist Systems in 2019 for $405 million, comes into play. At Juniper they used Mist technology to integrate artificial intelligence into all the components of the communication network, and allow communication providers and organizations to manage the networks more autonomously, especially when it comes to service assurance and automatic problem solving.

According to Manfred, Mist’s AI technology creates synergy with Juniper’s traditional hardware solutions. “Our goal is to bring artificial intelligence to the world of communication providers – to enable automation. however, Artificial intelligence requires data – and a lot of it. Data is not created by itself. There is a need for components in the network that will generate data, monitor latency levels, the rate of data packet loss, energy consumption, alert if the processors are overheating, and so on. The more information you have, the better the performance of the network will be. And here lies our advantage in the market, in the combination between hardware that provides the data and artificial intelligence that gets the best out of it.”

Manfred adds: “Together, this provides the communications operator with a holistic end-to-end view of the network, from the level of the optical cable and end device to the WiFi connection and frequency allocation. for example, A large communications operator that implemented our solution reported a 90% drop in user compliance. spoken In a dramatic reduction in the burden on the IT teams, and an incredible increase in customer satisfaction.”

The power goes to the edge

As mentioned, the new networks need to support not only human end-users, but many types of IoT accessories, some of which are responsible for mission critical applications. Manfred believes that only artificial intelligence can enable reliable supervision and monitoring of IoT applications. “Unlike a human user, an IoT accessory cannot complain. What we bring to the market is the ability to diagnose not only the experience of the human user, but also to monitor and understand the ‘experience’ of applications on the network, such as a smart traffic light or a smart city camera. The artificial intelligence makes it possible to monitor the performance of these applications based on parameters such as latency or fluctuations, to warn of problems and to drive proactive intervention.”

This need, to support critical applications that require zero latency, is what is pushing the network to be more distributed. For example, road safety applications, such as the field of communication between vehicles (V2X), cannot rely on communication with a remote cloud, but must be based on edge clouds, which are in fact local servers that allow the processing and routing tasks to be performed closer to the application itself .

The edge clouds become local communication nodes in what are called “metro networks” (MAN – Metri Area Network). According to Manfred, managing those metro networks is the next challenge for communication providers. “The intelligence is approaching the edge, mainly due to the need for zero latency. A decade ago the networks were centralized, and built around a few central pairs. Today the network is becoming more and more distributed, which requires much higher bandwidth and service security at the edge. According to estimates, traffic in the metro networks is expected to increase 5 times in the coming years.”

A few months ago, Juniper launched a new platform called Cloud Metro for the automatic management, from the cloud, of all the metro networks under the responsibility of the telecommunications provider. According to Juniper, the use of the platform dramatically reduces the maintenance costs of these networks and shortens the time it takes to add a new device to the network from hours to minutes.

Manfred: “For the telecommunications providers, this involves huge capital investments and a tremendous operational effort. We want to simplify this. The Cloud Metro platform allows operators to manage all metro networks like they manage a WiFi network, with each metro router being an access point. This allows them to manage the distributed networks automatically, remotely, in a software-based manner, and to improve service and reduce costs.”

Published in the categories: IOT, artificial intelligence, fifth generation, news, cloud, interviews, communication

Posted in Tags: Juniper


Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Recent News

Editor's Pick