What is Edge Computing?

Edge computing is a distributed computing paradigm where data processing and application execution happen at the network edge, typically on servers located within ISP points of presence (PoPs), cellular base stations, or dedicated edge nodes. Unlike traditional cloud computing that aggregates compute in a handful of large data centers, edge computing pushes computation closer to data sources and end users. This placement reduces the physical distance packets must travel, cutting round-trip time (RTT) from possibly hundreds of milliseconds to single-digit milliseconds for nearby users.

How it works: An edge node is a small-scale server running a container or virtual machine that executes application code. When a user makes a request, a DNS or anycast routing system directs the request to the nearest edge node. That node processes the request locally, possibly caching results or calling backend services only when necessary. Content delivery networks (CDNs) have naturally evolved into edge computing platforms; for example, Cloudflare Workers and Fastly Compute@Edge run JavaScript or WebAssembly on every edge server. AWS Lambda@Edge and CloudFront Functions extend serverless compute to CDN edge locations. These services handle tasks like A/B testing, URL rewriting, authentication, image optimization, and IoT data filtering without requiring a round trip to a central region.

Where it sits in the wider stack: Edge computing sits between the client device and the cloud. It is not a replacement for cloud computing but a complement. Latency-sensitive workloads such as real-time video processing, autonomous vehicle coordination, industrial control, and gaming require edge compute because waiting for a distant cloud is too slow. Edge computing also reduces backbone data transfer costs by handling data locally and only sending aggregated results upstream. The OpenFog Consortium (now merged into the Industrial Internet Consortium) and the ETSI Multi-access Edge Computing (MEC) standard define frameworks for deploying edge infrastructure, especially in 5G networks.