By installing a micro data centre on the site of Le Golf National, tournament organisers were given access to real-time data analysis and could help drive footfall to merchandise and food stalls
Edge computing is helping companies extract value from customers who connect to their wireless networks, by analysing data close to the source. Dan Robinson speaks to Aruba Networks EMEA boss Morten Illum about examples of the tech being used in settings as diverse as the Ryder Cup and University of Cambridge.
With its slick greens, vast rippling fairways and innumerable links-style bunkers, Le Golf National was an ideal stage for the 42nd Ryder Cup.
For the 270,000 spectators who descended on the 150-acre golf course just outside Paris in 2018, they were treated to a sporting spectacle as Team Europe regained the trophy from Team USA.
And operating behind the scenes trying to maximise their experience was Aruba, the networking subsidiary of Hewlett Packard Enterprise that created the wireless and wired connections to deliver key data insights.
By bringing its computing power to the “edge” – as in creating a micro data centre on site – it was able to use fans’ connections with the official Wi-Fi network to provide these in real time.
Edge computing examples: Ryder Cup 2018
“We had discussions about how to connect better with the fans and give them a better golf experience,” recalls Morten Illum, who runs Aruba’s EMEA business as vice-president. “It wasn’t just how to have a faster network but there was a business case too by using Wi-Fi as a tool.
“One way was using location-based services. At the grandstand you’d get a notification about the next player coming to the 18th hole, or you could see what the layout of the hole is when you’re stood at the eighth green and find where Tiger Woods is playing right now when you want to go watch him. They might also want to watch a live video of what’s happening on a different hole.
“If you have 50,000 people around an 18-hole course, the local network isn’t scaled for the massive amount of tweets and videos people are sharing and consuming. Even a 5G network would be limited and some people are roaming rather than on the Wi-Fi.
“So we wanted to provide a network that gives them those capabilities.”
Aruba claims to have achieved uninterrupted connectivity for 99,000 devices and accommodated 3.2 million Wi-Fi sessions.
It provided a total connection time of 63 years and facilitated 59 terabytes in data transfers, as well as a maximum of 15,460 connections made simultaneously.
But it also identified commercial opportunities by bringing spectators online.
Organisers noticed certain merchandise stands, along with food and drink vendors, would have long queues while others were quiet.
By feeding this data to users on an app, they could find the quieter stalls – thus distributing people across the site better.
Limited time special offers for goods in event stores could also be made at quiet times to stir footfall and ultimately increase revenue.
Illum adds: “It also helped us to understand how many people who buy a one-day ticket will actually come back for the other days.
“It gives us a better understanding for future tournaments about the offers we might make.”
What is edge computing?
Edge computing is one of the latest tech buzzwords that has found its way into the lexicon following the rapid expansion of the internet of things (IoT) – the system of devices ranging from mobile phones and smartwatches to coffee machines and cars that are connected via the internet.
Gartner estimates there will be 20 billion IoT devices this year, while Cisco predicts this number to hit 500 billion by 2030.
In conventional terms, data produced by these devices is sent via the cloud to a data centre – often managed by one of the tech behemoths like Amazon, which stores much of its hardware in California but also has sites across the world – where it is processed and sent back to the device.
The time this takes isn’t exactly lengthy but in some use cases, it’s not quite quick enough to achieve real-time demands – while there’s also been questions around how these companies use personal data.
That’s where edge computing comes in. The “edge” in this context refers to its literal geographic position as it is computing that’s done at or near the source of the data, often in some form of micro server or data centre.
It means the time taken to transfer and process data is next to nothing, reducing latency and enabling real-time answers to questions.
“It’s really about the next transformation going on in IT,” says Illum, who is based in Copenhagen but reports into a worldwide organisation headquartered in the US with 7,000 staff and $2.9bn revenue.
“The number of devices connected into the network is growing and in this new world, we’re creating an enormous amount of data.
“In order to use that data, it’s not much good transporting it large distances into a data centre.
“The best example is an electric connected car, where a company like Tesla is producing a huge amount of data each and every second that has to be analysed in the car.
“It can’t just send it to the cloud and back to say ‘do I need to avoid this cyclist?’ because you’ll crash into the cyclist if it takes too long.
“So Tesla has created a computer in the car that’s handling this data – and that’s what edge computing is.
“It exists in hospitality, where you want to have some capabilities of computing centres within the hotel itself.
“And it exists in hospitals where you need to analyse data quickly to give better patient care and have a good grip of whether you’re performing to standards.
“It’s located as close to the edge of a network as possible. In a campus environment, a company will have some compute power on the campus that consumes and processes data produced on site.”
What is the intelligent edge? Cyber security benefits
While edge computing can improve latency and address privacy concerns, for Illum it can go a lot further by enabling smarter decision-making.
“I wouldn’t even call it edge computing – it’s the intelligent edge,” he says. “You need to add intelligence into that environment to produce the decisions the data provides in some kind of smart way.
“You’ll quickly see that by having a cloud-enabled, data-driven approach – particularly with micro data centres moving to the edge.
“With the sheer amount of data coming in, CEOs need to embrace it because it can help them to be more efficient and productive.”
One benefit of the intelligent edge is in cyber security. Aruba sells a software package called IntroSpect, which uses advanced machine learning to quickly identify attacks before they cause damage.
While firewalls have been the preferred defence mechanism until now against malicious operators, Illum says the next generation of security threats will be active when users don’t realise they are under attack.
He gives the example of a worker accessing a café’s Wi-Fi to do some work, being infected by a malicious virus and then returning to their workplace to access the corporate network, where the bug reactivates.
“What IntroSpect does is, if your computer is behaving in an abnormal way by downloading HR files, it realises you don’t normally do that so it detects a threat,” explains Illum.
“So you can block those downloads immediately based on that intelligent edge in the system.
“The same could happen with a coffee machine. Every day it sends a data stream back to the vendor saying it’s sold 10 coffees, which is a small amount of data.
“If it starts downloading a 400MB file and sends it to an unknown IP address you’d know something is happening – so the intelligent edge can react to live threats.”
There could also be environmental benefits from edge computing by creating a sustainable office that saves energy.
Many offices can be lit up at night but a system could be installed for when there’s no one in the building that automatically turns down the lights, as well as heating and cooling.
“Aruba access points have that capability today,” says Illum, who wrote a book last year titled Opportunities at the Edge. “We can link devices to a sensor that turns down the lights and heating when people leave, and then when they arrive in the morning it automatically turns on again.
“There seems to be a strong push towards the edge by utilising some of the devices that are already there.”
Edge computing examples: City of London
Fewer than 10,000 people live in the City of London – the historic, single square-mile local government area in the UK capital – but each day, about 500,000 people commute inbound to work in one of the world’s largest financial hubs.
With so many people crammed into such a small space surrounded by huge towers, it poses a significant challenge in creating a robust network.
In a bid to introduce flexible working, the City of London Corporation, the local authority, employed Aruba’s network architecture expertise to help it rolls out Microsoft Office 365 to 4,000 staff working across 120 locations stretching from the Square Mile to Heathrow in the west and Epping Forest in the east.
It has enabled 70% of staff to work from mobile devices and transformed productivity with a greater use of mobile applications including video calls, digital stock control and even remote fingerprint scanning in the case of the City’s own police force.
Despite creating a single network that consolidates the police and corporate operations, it retained a clearly defined, separate network infrastructure for management and security.
More than 1,000 Wi-Fi access points were installed across the 120 sites, with computing power stored “on the edge” nearby to monitor and manage performance and availability of the network and applications.
Illum explains: “The business case was about collaboration and modernisation – giving everyone access to the same professional environment, whether they’re working in the City or outside.”
The City has big plans for using the network in the future.
Sean Green, director of IT for the City of London Corporation and City of London Police, says: “Undoubtedly there will be opportunities in the future to do much more with mobile applications.
“There’s an opportunity to incorporate IoT, possibly around smart building solutions or air quality sensors.
“We’ll be better at monitoring, managing and proactively improving the City environment.”
Edge computing examples: University of Cambridge
Dating back to 1209, the University of Cambridge is one of the oldest and most prestigious educational institutions in the world.
Over the past 800 years, it has gradually sprawled to now occupy 1,000 buildings across the city, and for its 18,000 students, getting around can be something of a challenge.
Aruba has installed 4,500 wireless access points to establish a single network for much of the university that helps with onboarding by enabling students connected to a Wi-Fi network to navigate buildings and find classrooms, lecture halls, food areas and shops on campus.
The network now accounts for a quarter of the university’s annual internet traffic, up from 4% in 2012.
Edge computing adds a layer of intelligence to the network by analysing behaviours on campus, such as how personal devices are used, which common areas are used the most and even spending habits in pubs and coffee shops.
It can also keep the network up to scratch by identifying likely failure points and has achieved a failure rate of less than 0.1%.
Previously, the 31 colleges each had their own IT strategy and budget, creating a disjointed service, but at least 21 of those have now joined Aruba’s network.
There are also plans to install access points in parks and public spaces, using edge computing to enhance the visitor experience by helping people find museums and other attractions while analysing their behaviours.
Illum points out how it also provides extra security for intellectual property rights – a key area for a university whose research prowess is regarded as world-class.
“Cambridge has a lot of research so we’re protecting what that research is in a controlled network,” he adds.
“We want everyone to be on the same network so we can secure it in a smart way as threats become more powerful.”