AI and Machine Learning: Allies in Digitalisation and Improved Network User Experience

By Jamie Pitchforth, Head of UK Strategic Business at Juniper Networks.

  • Posted in

AI is one of the technologies that has advanced significantly in recent years as a result of changing market conditions, and received a lot of attention, particularly in the IT industry. AI is already producing measurable advantages, from proactive support for self-driving network optimisation to much-needed user experience knowledge. The idea of AIOps (Artificial Intelligence for IT Operations) is increasingly gaining traction from customers and providers, despite the fact that AI and machine learning (ML) are already widely used.

AIOps combines big data and ML to automate, control and optimise processes and IT operations. One of the key targets of AIOps is to reduce the Mean Time To Resolution (MTTR), providing a set of functions able to identify the root cause of existing problems, suggest possible solutions and automate the operational activities.

Real support for digitalisation

Today, it is no longer sufficient for networks to merely function properly if we want to provide end users with a good network access experience. Instead, it's critical to gather information, insights, and analytics about not only the network infrastructure that allow users to access network services, but also the actual experience when using that service. It is impossible to comprehend when an issue arises and how to fix it without data and insights. Also, even when an anomaly in the customer experience is detected, it becomes fundamental to have tools and platforms that leverage AI mechanisms and ML algorithms. They must be capable of automatically defining the root cause and providing a path for automatic resolution by always delivering the best possible experience to the user.

By gathering quality data and leveraging ML algorithms, the AI applied to IT operations continuously monitors the network and processes data (coming from the devices and the users) in order to find performance and quality indicators within a device or connection. This information is then made available to the network operators who are thus able to immediately identify when a deviation occurs and to fix it even before the user is aware of the anomaly.

Best of AIOps and user experience

AIOps is an industry term that describes technology platforms and processes that enable IT teams to use AI to make faster, more accurate decisions and respond to network and system incidents more quickly.

AIOps contextualises massive amounts of telemetry and log data from the IT infrastructure of an organisation in real-time or almost real-time. It then combines it with historical data to produce useful insights. AIOps is a representation of an assistant with in-depth knowledge of the network and IT environment and the capacity to use that information to provide real-time analysis and carry out or suggest future steps.

AIOps increases the efficiency and performance of individual applications and services. Organisations using AIOps as part of their automated infrastructure and operation workflows are improving

security, outage incident response times and infrastructure purchases. Those just starting with AIOps see it as an investment in performance analysis, anomaly detection, and event correlation that gives them the ability to predict future network-impacting events.

The road to the AIOps Platform

Starting from data collected thanks to telemetry sensors and log systems, an AIOps platform can quickly identify not only the cause of any occurring problems but also the users impacted by them. This is exactly what Mist AI by Juniper Networks does by providing an “intelligent” interface which can be queried in natural language. What makes this possible is the AI engine Marvis, the heart of Mist AI. The Marvis Virtual Network Assistant (VNA) brings conversational AI to IT operations. It continually learns about the network as it ingests data and contextualizes requests for automated, predictive actions. It continues to build out its Natural Language Processing (NLP) capabilities with Natural Language Understanding (NLU) and Natural Language Generation (NLG) so users can interact with it as if it were another team member.

These days, network performance and connectivity are crucial. In order to make the user experience as seamless as possible, network management should be streamlined and access and performance issues can be decreased. A "basic" SD-WAN can be transformed using Mist AI technology into an AI-driven WAN that prioritises the user experience. Mist AI uses AI and ML to simplify the network and automate a number of crucial tasks. With automated and remote troubleshooting, issues are proactively fixed before users even become aware of them. Delivering enhanced network experiences in this way will change the way people connect, interact, and live.

Procurement expert Keith Hausmann highlights how NLP and machine learning are driving efficiencies and savings while enabling sourcing teams to add more business value.
By Peter Greiff, Data Architect Leader EMEA, DataStax.
Keerthi Mukku, Commercial Account Manager at Plextek investigates the importance of Foreign Object Detection (FOD) at any airport.
Josip Rozman, AI consultant at Plextek examines if we should be using AI to write our content.
By Rohit Bhosale, Persistent Systems.
By Kirsty Biddiscombe, UK Head for AI, ML & Analytics at NetApp.
By Mark Lowe, business development director at North.
By Howie Xu, Vice President of Machine Learning and AI at Zscaler.