Edge Computing
Edge computing is processing and analyzing data close to where it is generated, at or near the edge of the network, rather than sending everything to centralized cloud servers. Keeping computation local reduces latency and bandwidth use and enables faster decisions for IoT devices, cameras, and real-time systems.
In an edge setup, a small server, gateway, or device sits on site next to the equipment or sensors. It runs the software or AI model locally and sends only results, alerts, or summaries to the cloud for storage, reporting, and model updates.
A common example is vision AI on a production line: cameras feed an on-site edge device that inspects each item in milliseconds and can reject a defective part immediately, without waiting on an internet connection. Roadside cameras that detect overheight vehicles or flooding work the same way.
Edge computing complements cloud computing rather than replacing it. The cloud is better for heavy model training, long-term storage, and cross-site analysis, while the edge handles time-sensitive decisions where the data is created. Edge AI is the specific case of running AI models at the edge.