Azure Edge Zones – Microsoft’s Push to the Front of Edge Computing

Thursday, May 7, 2020

Businesses and, industry, in general, are relying more and more on cloud technology to improve their operations, security, and communication. Awesome as it may be, the cloud’s processing power is often only as good as the geographic distance between you and the location of the server. This is exactly the reason why Google Stadia is a monumental flop. For now.

To give you another example, if you want your security systems to distinguish between employees and newcomers, you can’t wait for that information to be sent to a datacentre, analysed, and then responded to.

Similarly, if want to automate your warehouse, you will need an on-site solution that can do the task and deliver a result to anywhere on your premises within milliseconds, instead of a terminal whose job is to send and receive data over the Internet, like Alexa or Google Assistant.

Azure Edge Zones is Microsoft’s solution to this problem. If you want to automate parts of your operations and are in need of on-premise data processing, Azure Edge can extend the functionality of your (hybrid/multi) cloud with small-footprint extensions of its network.

Image Source: microsoft.com


What is Edge Computing?

Don’t get me wrong. We live in a time of high-speed Internet connections where most developed and developing countries have access to low-cost, high-performing Internet infrastructures, which (by the way) continue to expand. The problem is that intelligent processing, such as the kind you’d need to coordinate thousands of load-carrying robots in a warehouse, requires near-instant, under 30 milliseconds, communication.

At the same time, you might want to grow or expand your businesses to geographic areas that are quite remote from major data centres – think of rural communities in general, but also more specific examples such as Africa and its emerging economies, most of Russia and China, the Caucasus, Central Asia, and more. Edge computing refers to intelligent processing that takes place where the data originates, at the source.

With high-performing processors, edge computing is extending the functionality and reach of the cloud, so that you’re always close to it, even when you’re far away from high-speed networks and massive data centres.

For instance, if Google manages to figure out how to reduce the network latency due to communicating over great distances, its Stadia will become a true wonder. One way or another, they’ll figure out, but they’ll likely use edge computing as part of the solution.

Image Source: microsoft.com


What Are Azure Edge Zones and How Do They Work?

Microsoft has a definitive plan to take the pole position in edge computing with the help of 5G. As a result, they’re releasing no less than three edge computing services to help businesses who are far away from Azure regions:

  • Azure Edge Zones, which lets your run virtual machines, containers, and a number of throughput-intensive applications without having to worry about communication delays.
    Use cases include streaming and dynamic content delivery, gaming, but also real-time analytics that need to interface with machine learning and AI.
  • Azure Edge Zones with Carrier, which are placed in mobile operators’ networks to ensure near-instant latency (less than 10 milliseconds) for applications that also run on mobile devices. If you need to be close to your customers or to your on-the-go employees, this is the way to do it.
    To previous use cases, Edge Zones with Carrier adds things like tele-medicine and connected cars, vans, or trucks. In the US, this option is offered in partnership with AT&T.
  • Azure Private Edge Zones, which are the on-premise version you need to deploy software created by you or independent vendors, as well as run virtualized network functions. With this option, you’ll be able to run private networks that also extend over mobile infrastructures (LTE and 5G) and manage them with the help of Microsoft’s cloud technology.
    Additional use cases include geo-fencing and firewalls, real-time command of production lines and robots, machine vision, real-time analytics, but also enhanced surveillance.


Image Source: microsoft.com


These solutions are helping industry bridge the gap between 5G and the cloud in order to create a streamlined, low-cost, high-performing cloud experience at the edge of networks. With Azure Edge Zones, you’ll be able to connect your remote users/branches with the cloud, but also to one another.

The true innovation comes in the Private solutions, which allows the Azure Stack Edge hardware to extend its connectivity through mobile networks in order to guarantee near-instant processing and connections for mobile devices as well. Basically, the more Edge Zones are functional within an area, the better the coverage.

Traditionally, the major problem with including carrier infrastructure in edge computing has been the trombone effect. Tromboning is a by-product of the practice of hosting information in far-away data centres and what it means is that, instead of getting a piece of content you want from your premises several miles away, your carrier will first send it to the main data centre (which may be tens, perhaps even hundreds of kilometres away) and then send it back to you from there, much like the way sound waves travel in the eponymous musical instrument.


Image Source: microsoft.com


Fortunately, Azure found a way to avoid tromboning – to the extent that carrier infrastructure allows it. In the future, they plan to enable individual developers to distribute workloads wherever they might need them on the edge using more sophisticated code placement. As you probably expected, Google and Amazon also have plans to roll out similar services, which means that edge processing and local connectivity is going to get a lot of attention in the near-future.

What Does the Future Hold?

The best part about integrating the cloud with LTE and 5G is that on-premise, independent developers, or independent software vendors, don’t need to have engineering expertise in carrier infrastructure to leverage it. In the case of Azure, Microsoft and its local partners will set up the network itself, from the boxes to the firewalls, vLANs and everything else you can think of. This will enable businesses to employ high-level programming constructs and orchestration layers to automate manufacturing and warehouses.


Image Source: microsoft.com


Overall, it means bigger savings for businesses and major industries. Azure Edge Zones can be used in tandem with AI and machine learning to employ real-time analytics because they allow data from your IoT to be processed on premises before being sent to your corporate data centre and/or the cloud. For example, with the help of industrial cameras, you can have a setup that identifies product layout flaws or even tells the difference between the scratches on a beer bottle and possible liquid contamination.

Whether you’re trying to save money by coming up with an intelligent setup on your premises or attempting to bring more complex services to your customers, it’s likely that edge computing of the kind that Microsoft is rolling out will lower overheads and enable companies to focus more on innovation.