One of the biggest growth areas in IT right now is edge computing, a framework that sees organisations bring computing and storage power out of the data centre and towards the places where data is generated and stored.
By the end of this year, there will be an estimated 55 billion edge devices on the market, with this number expected to nearly triple by 2025.
The rise of edge has been driven primarily by more data-intensive workflows, particularly through the widespread adoption of AI and 5G.
The appeal of the edge is simple: by cutting down latency between the collection and processing of data, it promises to deliver faster and more reliable services to organisations and customers. For this reason, adopting the edge is a top priority for many businesses across many industries.
However, a transition to the edge represents a complex change to your IT infrastructure. To make an edge transition work, you need to ensure you have some crucial prerequisites in place that lay the groundwork for a successful move.
From the outset, you’ll need to have standards built on open technologies, leverage the hybrid cloud, and ensure your edge environment is built to scale.
Standards built on open technologies
Edge computing relies on enabling varied and far-flung pieces of equipment talking to one another seamlessly: whether these be storage racks, compute nodes, or sensors at the frontier of an edge network.
This in turn, however, means that your edge environment has to draw on a diverse range of suppliers and configurations in order to be practical.
To guarantee all devices in your edge ecosystem can talk to one another, they need to share a common language.
And the best common language available to edge ecosystems is that of standards around open source software and hardware, as it means that you no longer have to worry about the provenance or minutiae of your edge devices, so long as they are compatible with whatever open standards you employ.
The hybrid cloud
Given the aforementioned diversity of components at play in an edge network, you can also expect a wide variety of workloads in your edge infrastructure.
For example, you may find yourself running virtual machines, containers, and bare-metal nodes at the same time.
To allow these disparate workloads to operate in harmony, you need to deploy an infrastructure that fully integrates your public and private cloud environments into one unified experience.
This is what is called the hybrid cloud, and it represents the most common foundation for an edge ecosystem used to manage many thousands of diverse devices and processes.
Attention to scale
One of the biggest benefits you’ll find from edge computing is its ability to scale – both in terms of the variety of workloads, the number of devices, and the geographic range it covers.
But this benefit can only be accessed if an organisation creates their edge infrastructure with the intent of scaling it up.
To achieve this, an edge environment should be structured and planned to accommodate new technologies that may be incorporated into it, and also be able to anticipate new workloads and user categories.
One example of good planning and attention to scale can be found in security planning: it will always be easier to plan your permissions system ahead of time, instead of having to consistently rewrite and replace an ad-hoc structure that’s not fit for purpose.
Through adopting open standards, building atop a hybrid cloud, and having scale front of mind from the outset, your organisation can deploy edge computing with the confidence that it will be reliable, scalable, and effective at solving whatever business challenges you may face.
by Martin Percival, Solutions Architect, Red Hat
About the Author
Martin Percival manages Red Hat’s Solutions Architect team, helping others understand Red Hat’s growing technology offering and the Open Source model upon which it is based.
Having worked in the IT world for 36 years, Martin’s love of explaining IT to varied audiences has seen him specialise in bringing those explanations to customers, the press, and analyst communities.
Article by [author-name] (c) Irish Tech News - Read full story here.