by CIO Staff

Building an AI-driven network

Sep 18, 2019
Artificial IntelligenceEducation IndustryGovernment

Bob Friday
Credit: David Thompson Photography

Artificial intelligence (AI) – it’s a nebulous term that means many things to different people. What is true is that one day in the near future, machines will be likely to possess ‘human-level’ intelligence, providing organisations with efficiencies that they have never seen before.

But what role is AI playing inside organisations today, particularly when it comes to providing a good experience for internal users and external customers across their wide area networks?

Tech execs gathered in Sydney in September to discuss the benefits of using artificial intelligence technologies inside their wired and wireless networks.

Bob Friday, chief technology officer at Mist, a wireless networking equipment provider which Juniper acquired in March this year, said that an AI-driven network maximises the user experience through better performance and reliability while lowering IT costs through better efficiencies.

“For example, if something goes wrong, our AI-driven virtual assistant can easily pinpoint the source of the problem. [It will ask] was it a network, device or application issue? There is no need to hunt through event logs or dashboards or send anyone onsite with handheld devices to try and recreate problems,” said Friday.

Mist also proactively notifies IT departments of problems before users often know they exist and enables full automation of help desk tasks.This eliminates the time and money previously spent troubleshooting networks.

In fact, one Mist partner estimated up to 40 per cent OPEX savings with AI, he said.

“But it is more than just management or troubleshooting. Mist is currently the only networking infrastructure vendor utilising its own AI and machine learning (ML), to support all our customers fundamentally changing the paradigm from providing enterprise IT network with some management software – to helping the manage in real-time the end-to-end mobile user experience of their customers, employees and mobile devices.

“For enterprises managing critical services, we have seen that this is not just a key benefit, but one which is very quickly becoming a mandatory requirement,”he said.

Friday added that AI is only as good as the data collected.

“To that end, Mist collects over 150 user states in near real-time that give visibility into the status of a wireless user’s experience. This is combined with data pulled from Juniper switches to create wired and wireless service levels, he said.

“For the first time in the industry, you can actually set, monitor, and enforce performance goals for key metrics like throughput, latency, and roaming time and measure the exact impact they are having on the user experience.

“As we say at Mist, ‘Up is not the same as good.’ This means that just because the network is operational doesn’t mean users on the network are having a good experience. By delivering a more user-centric model using AI, Mist lets IT focuses on delivering the best user experiences. “

Joe Perricone, chief operating officer, at Swanson Health Asia Pacific, said his organisation is using AI-driven network applications to assist in monitoring of front and back end infrastructure services. This allows a direct connection and proactive approach with its customers and users of its systems, he said.

“Furthermore, the reliance of our AI is critical to the performance of our manufacturing and job dispatching processes, as well as the efficient delivery of goods and services in the field,” he said.

“This is achieved by creating intelligence connections with our key applications.”

Perricone added that using the power of AI, the organisation is able to capitalise on faster delivery times and routes.

Swanson Health’s mobile workforce can proactively report on issues, and tasks are automated immediately to the right person or prioritised where everyone has access to all the information required to complete tasks un a unifed and collaborative manner, resulting in better use of resources and cost savings, he said.

“Our marketing efforts involve sending more specialised communications at a higher frequency, targeting specific customer segments with specific product ranges,” he said.

“This makes it difficult to do A/B tests manually. It might take a few weeks to run a single test so doing this for hundreds of product ranges is a considerable task. Our marketing application needs to be able to monitor the performance of the communications being sent and make adjustments accordingly to improve the communication and performance of our teams.”

Meanwhile, Adam Rendulic, director of IT, ANZA at materials manufacturer, Kion Group, said his organisation is not yet using artificial intelligence across its networks but it has ‘some AI’ or predictive/adaptive learning capabilities within its storage array.

“This is beneficial in highlighting concerns or areas that are not optimally configured and ensuring bottlenecks are addressed. Also, if a pattern is detected within the customer base, it will then look for other customers with similar parameters and inform them of the issue before it manifests,” he said.

Russell Morris, chief information officer at TransGrid, said the electricity network operator has started developing AI-driven networks and has subsequently deployed an SD-WAN solution which is being resold to a telecommunications business.

“SD-WANs are more resilient to faults and have some self-healing capabilities, leading to improved customer uptime,” he said.

Morris added that a combination of factors are used to deliver a better user experience.

“Linking our service desk and self-service [capabilities] with ServiceNow allows the equipment to log faults proactively so we can spot trends early and maintain service levels,” he said.

Automation and the question of job losses

As with any automation activity, adding AI capabilities to the network can strike fear amongst staff that introducing the technology will result in job losses.

IT skillsets always need to evolve and AI is simply another tool that needs to be leveraged as an organisation grows and evolves into the modern era, said Burke.

“With AI, IT saves time and money, optimises user’s experiences and maximises available resources. Staff can shift from many, rote tasks to projects that deliver more value for the business.”

Andrew Atkin, head of information technology Asia Pacific at chocolate maker Lindt Sprungli, agrees.

“For AI, machine learning and automation technologies, there is no consideration of job losses. With the removal of manual and repetitive tasks, people can concentrate on the value-added and analytical tasks to bring real value to the business.”

Swanson Health’s Perricone said the organisation has not had to decrease or increase staffing levels due to the effectiveness of AI technologies.

“We have found the use of AI technologies not only enables better cost management, it also provides more room for quality control, resulting in better outcomes for our customers.

“We need to develop and evolve further data and AI-driven workflows that can help us understand our customers better, predict trends in the market and adapt accordingly.”

Mist’s Friday added that IT departments are increasingly recognising AI and ML as an enormous benefit, and not a threat.

“These technologies provide insights not previously possible, especially as the volume and complexity of devices continues to increase. In fact, it is increasingly crucial in order to be able to keep up with the complex requirements of the smart device era,”he said.

“They can now utilise resources more efficiently and offload the mundane, manual tasks such as finding hardware problems, bad cables or software bugs. Instead, teams can now better focus to better focus on business-critical projects such as customer engagement analytics, customer experience, facilities management and optimising business operations.”

Friday said AI also provides much needed automation that reduces errors from manual configuration.One Mist customer, he said, moved from deploying three new sites/week to 300 new sites/night with Mist.

“IT skillsets always need to evolve.AI is simply another tool that needs to be leveraged as organisation grow and evolve into the modern era. With AI, IT saves time and money, optimises users experiences and maximises available resources – i.e. IT staff can shift from manual. rote tasks to projects that deliver more value to the business,”he said.