Monitor edge latency for IoT devices
Latency in edge devices can have several sources, stemming from software, network issues or old hardware. Here's how you can monitor and troubleshoot latency at the edge.
Monitoring edge devices for latency enables you to track and identify issues before they become bigger problems. Most companies already have latency monitoring tools in place; it's just a matter of using them to track edge devices, too.
IoT technologies such as edge AI and fog computing enable you to support a more decentralized network and operating model. As you become more dependent on edge computing, however, you might notice issues with edge latency, especially for networks that use legacy technologies or those experiencing exponential growth of IoT devices. The network sometimes simply cannot handle the added work, so bandwidth and system performance suffer.
Remote and network monitoring tools can identify edge latency issues and help IT teams be proactive in fixing them.
Latency in edge devices
Generally speaking, latency happens due to the physical distance between devices. Some companies might have thousands of kilometers between an edge device at a manufacturing location and the system monitoring it. That's the double-edged sword of edge devices: You can deploy them easily, but they might be far away from the systems they send data to.
Other reasons for edge latency include:
- Data management bottlenecks as devices decide what data to keep, what to transmit and what to discard before sending it downstream. This can create latency if devices process massive amounts of data.
- Device update delays if edge devices infrequently connect to the internet or the network to transmit. You might prioritize updating the device first, but this can cause delays in other functionality.
- Legacy servers and network architectures that aren't up to the challenge of edge devices and struggle with the increased bandwidth, usage and performance needs.
- Cybersecurity software that struggles to oversee and enforce identity, authorized access and other security policies in edge deployments. Trying to impose the same rigid security from a data center on a different location -- such as an industrial space or retail store -- can introduce latency to edge devices, because such devices might not be always-on, might have intermittent internet connectivity or might be deployed in public areas.
You can avoid latency with edge devices by using a content delivery network (CDN) to manage edge applications and devices. CDNs can send video by connecting devices with video-specific data centers and services to reduce latency. They can also centralize and host the static portions of web apps that connect to edge devices and increase app performance that way. CDNs also reduce latency by hosting and distributing firmware and app updates in a decentralized way.
You can reduce edge latency by deploying network or edge monitoring software. By monitoring the edge devices and connected systems, you can better understand device availability, track when and how devices connect and identify bottlenecks and performance gaps. You can then deploy updates and fixes more quickly and efficiently.
It might be difficult for you to get to the physical edge of the network to check devices and fix latency issues, which means you should consider remote monitoring and management applications. These applications offer clear overviews of devices, alerting and reporting, which keeps you updated on the device.
Monitoring software such as SOTI Connect, Varonis Edge, TeamViewer Remote Management and PRTG Network Monitor track and alert you to changes on edge devices. They can track everything from CPU and RAM changes to patch and update notifications. Setting up an edge performance strategy can help, too, because it outlines the baseline levels each device requires and the system and business process each device supports. For example, devices handling security locks should take higher priority than those tracking mileage for delivery vans.
Monitoring can help you move beyond simple edge latency alerts and reporting and enable you to optimize edge performance. For example, you can use latency data to:
- identify unnecessary connections and usage;
- simplify device connections and priority;
- pinpoint the source of a cloud service bottleneck;
- reduce data storage costs by moving relevant data storage to the edge;
- optimize network system performance;
- design new edge use cases or scenarios; or
- choose a new connectivity package for the edge device -- for example, move from Wi-Fi to LP-WAN or Bluetooth.
Troubleshooting edge latency
Latency issues in edge devices often involve a combination of online and in-person work to fix. The first step is to try online- or digital-based fixes, such as a ping or traceroute testing. Other tests to try include WAN performance test and trace, network connection tests, internet connectivity tests and over-the-air latency testing if the device connects wirelessly.
If the latency issue arises from the device itself, you might need to dispatch an IT technician to the site to verify the problem. However, most latency issues come from software or other digital issues that you can resolve remotely.
Edge devices are only useful if they're as responsive as companies and businesses need them to be. Even 3 milliseconds of latency can create issues for some industries, so learn how to avoid and track latency at the edge before it impacts those devices.