Image generated by AI: © Linus/Stock.adobe.com
Red Hat’s Mark Swinson spoke to SiliconRepublic.com about how edge computing has evolved in recent years and the transformative impact this technology is bringing to cities, transport and the military.
Edge computing is an emerging tech sector that has been growing in maturity, with more companies focusing their attention on this space.
Last month, Dell announced an upgrade to its customer solutions centre in Limerick, in order to include more forms of emerging tech including edge computing.
The concept of this technology involves bringing the processing power of computers closer to data sources and data consumers. Mark Swinson, enterprise automation specialist with Red Hat, says this is particularly useful when “highly responsive behaviour is required”.
“Overall, edge computing enables distributed and decentralised processing, bringing power and intelligence closer to where it is needed,” Swinson said.
The technology is generally associated with the internet of things (IoT), which is the embedding of sensors and processing capabilities into everyday objects. But as the technology advances, edge computing can also be used to boost research in various sectors.
Advances in edge computing research
Swinson said the technology requirements of edge are similar to those of centralised data centre processing systems, which means the focus is on more powerful chips, lower power consumption and “more flexible deployment of workloads”.
But as the computing resources are deployed outside of data centres, there are other requirements such as a low-touch or no-touch set-up and physical security. Thankfully, technology advancements around the IT sector have benefitted edge computing too.
Swinson said that advanced, self-powered sensors are now capable of collecting data at a lower cost, while more powerful algorithms make it possible to act on larger amounts of data and help users “make decisions with better insights in real time, with better accuracy”.
To further develop this sector, Swinson said Red Hat has made enhancements across its own portfolio, such as new update mechanisms that can roll back in the event of a failure, “reducing the need to send an engineer out to a remote site”.
One of the biggest challenges to tackle in terms of edge computing is being able to maintain consistency across “multiple compute landscapes”, according to Swinson.
“Increasingly, it’s important to be able to flex deployment – from edge to centre and back to edge,” Swinson said. “For example, collecting data to train an AI model from sensors at the edge, training a model in a data centre and then pushing the model out to the edge to act on data, while harvesting enough data to validate and refine the model.”
Edge computing breakthroughs
Swinson said that one key sector that has benefitted from the use of edge computing is the concept of smart cities.
“An IoT-informed digital-twin smart city platform brings a city the ability to gather relevant data from multiple sources and extract valuable insights, driving smart decisions for the benefit of all citizens,” Swinson said.
“The biggest challenge is to effectively gather and process large-scale data while, at the same time, building an easy-to-deploy solution that is scalable and secure.”
To help make this possible, Red Hat partnered with FIWARE, a German-based non-profit that is focused on building an open, sustainable ecosystem around public software platform standards.
In 2022, FIWARE partnered with the Red Hat Social Innovation Program to build an open-source, eco-smart city platform “that cities of any size around the world, can implement”.
“More than 250 cities around the world have adopted FIWARE standards for the implementation of their smart city strategy,” Swinson said.
Edge computing has also been used for sectors such as the military and transport. Swinson said Lockheed Martin has been collaborating with Red Hat to bring AI technology to “geographically constrained environments” for the US military.
“The companies are equipping US military platforms, such as the Stalker unmanned aerial system, with advanced software that was previously too large and complex for these systems,” Swinson said.
“This advanced software enables small platforms to handle large AI workloads, increasing their capability in the field and driving faster, data-backed decision-making.”
Red Hat has also worked with Swiss Federal Railways (SBB) to bring intelligent systems onto the country’s rail network. These systems include dynamic LED information displays, digital seat-booking systems and Wi-Fi access.
“However, managing the devices supporting these features was difficult due to high volumes and a lack of central control,” Swinson said.
“Red Hat’s Ansible Automation Platform helps SBB automate complex deployments and centrally control its IT infrastructure through a visual dashboard with features such as role-based access, scheduling, integrated notifications and graphical inventory management.”
10 things you need to know direct to your inbox every weekday. Sign up for the Daily Brief, Silicon Republic’s digest of essential sci-tech news.
