If you’re looking up analytics for your security system, you might hear the terms “on the edge” / “at the edge”, or “edge analytics”. Understanding the difference between edge analytics and other analytics can seem technical, even overwhelming. In fact, it’s pretty simple - so let’s get started.
Edge analytics are analytics applications that function within a device. They do exactly the same job as cloud or server-based analytics; the only difference is where the data analysis takes place. For example, many security cameras have in-built video analytics - we’d refer to these as “edge analytics”, because this embedded technology analyses data as the camera records it.
This technology isn’t confined to security cameras. All kinds of devices employ it - pretty much any internet-connected device within the Internet of Things (IoT) uses edge computing.
To understand why edge analytics are so popular, let’s compare them with server-based analytics:
| Edge analytics | Server analytics |
|---|---|
| Data is analysed within the device almost as soon as it is captured, minimising latency | Data is sent from devices to the server for analysis; over a vast network this can take time, increasing latency |
| Uses bandwidth more efficiently as data is not sent back and forth | Uses more bandwidth as data is sent back and forth between remote devices and the central server |
| Less strain on bandwidth is also less expensive | Transmitting data such as video or audio requires high bandwidth, which can be expensive |
In brief, edge analytics are light on bandwidth and more efficient, as analysis is decentralised and can take place almost instantly. They are especially useful if you want to analyse bandwidth-heavy data such as audio or video. What’s more, transmitting this kind of data over long distances also runs the risk of it degrading in transit, which means the information that reaches the server for analysis isn’t the same quality as it was when it was first captured. Edge analytics eliminates this risk altogether by analysing the data straight away - then sending the results to a central location.
You can read more about data degrading over long distances in our article on NVR vs DVR networks.
As we discussed above, edge analytics certainly have advantages over centralised, server-based analytics in terms of speed, data quality and bandwidth used. So when are businesses most likely to use them?
Edge analytics have many advantages, but as with any technology, they aren’t a one-size-fits-all solution. There are instances where a different option might be better. Let’s look at some examples where they wouldn’t work so well.
You can learn more about how to protect your camera network from hackers in our guide to Video Surveillance Cybersecurity:
For some companies, investing in edge analytics is a no-brainer, since the efficiencies it brings are worth the cost. However, it is crucial that you weigh up the benefits of using edge analytics against the considerations above - your business might not need them, or might not be ready for them just yet.
For those businesses that might struggle to justify the expense and upheaval of upgrading for edge analytics, there is another option: cloud-based analytics. A cloud-based solution offers all the speed, flexibility and scalability of edge computing, but at a more manageable price point.
Cloud-based solutions are often designed to integrate with a range of different systems, meaning it is possible to upgrade your legacy IT system with the help of software, rather than replacing all your hardware.
So how does it work? In many cases, a cloud-based solution sits between your remote devices and your central server, intercepting data and analysing it before it reaches your server. This solution can be rolled out across a handful of devices, or thousands of them; the cost to you remains relatively low, since you are only paying for a service, and not for devices, bandwidth, storage, and all the other network trappings.
Whether you use server-based analytics, analytics on the edge, or a cloud-based solution, your key consideration should be how it benefits your business. If you are considering upgrading your system to benefit from analytics on the edge, perhaps try a cloud-based option first to see if it really does improve your processes in the way you anticipate. You might find that it gives your current system a new lease of life for several for years to come.