How artificial intelligence will affect the future of networks – and what you need to do about it now
"Prediction is very difficult, especially if it's about the future," said Nils Bohr, Nobel laureate in Physics, and that observation certainly applies to evaluating technology trends and their impact on the business networks of tomorrow.
Today’s network is built to deliver applications and services that didn’t exist 15 years ago, and is designed for companies that may themselves only be a few years old and delivering products and services on devices that are changing the way we live and work.
Consider businesses of today. They face a number of different challenges and our role is to help them decide on the best approach and the technologies that can meet their needs. We can’t predict what the future will look like, but we can look at the latest trends and evaluate how they apply to the challenges our customers are facing today and those they’re set to face tomorrow.
It is appropriate then that the future network is intelligent, using Artificial Intelligence (AI) to spot problems before they impact businesses and is able to adapt to demand without requiring human intervention. In fact, Gartner states: “Your business context nurtures the use cases that are most pressing, most important and most urgent for your organisation’s business outcomes. For AI, start with the use cases before you — with what you, as an organisation, are striving to accomplish. Once you identify and scope the business problem, then you can identify the required skills, the data requirements and the relevant AI techniques.”
AI is a mechanism for making key decisions in a network – while it’s not the futuristic vision of a fully intelligent manager (at least not yet), AI can be trained to address specific problems, recognise patterns or to spot early signs of a fault or cybersecurity attack.
AI is at the core of the future network, helping manage unpredictable bandwidth requirements and virtualised network functions, reducing the amount of interaction from network managers and helping automate time consuming processes.
Why your future network needs AI
As networks become more complex and unpredictable, AI can be used to automatically manage this complexity without impacting the end users. Bringing AI into the network takes some of the responsibility away from network managers, freeing up their time while adding a level of oversight and intelligence that would otherwise be impossible to match.
One good example of AI in action is spotting the traffic patterns in the build up to a DDoS attack – by training AI to recognise the earliest signs of an attack, capacity can be automatically scaled up to deal with the increased traffic and then reduced when the attack is over.
Another one is the ability to use AI to predict bandwidth demand for a retail business, based on existing network data. The network could scale up to cope with seasonal sales or busier days, then drop down to a lower level in quiet periods.
AI can also help with fault prevention and fault finding – it can predict the risk of a component failing and advise network managers on when to act. It could also help identify faults much quicker than a human, particularly for applications in a virtualised environment.
For organisations collecting large amounts of data, AI can also be used to quickly interrogate data from multiple sources, providing an output report that would otherwise take days or months to deliver.
We’re already seeing AI in action in a number of the consumer devices we use every day. Soon, its role in the network is likely to become just as commonplace.
Interested in hearing industry leaders discuss subjects like this and sharing their use-cases? Attend the co-located IoT Tech Expo, Blockchain Expo, AI & Big Data Expo and Cyber Security & Cloud Expo World Series with upcoming events in Silicon Valley, London and Amsterdam and explore the future of enterprise technology.
- » A blueprint for open source language automation: How enterprises and developers can benefit
- » Why a risk-based approach to cybersecurity can save time and money
- » The big enterprise security problem posed by small – or at least smaller – businesses
- » Humanising automation: Understanding the five senses of artificial intelligence
- » The three biggest challenges facing the equipment-centric enterprise