Edge Computing

Edge computing is a distributed computing paradigm in which computation and data storage are placed close to the sources or consumers of data — at the network edge — rather than concentrated in a small number of centralized cloud regions. The aim is to reduce latency, conserve backbone bandwidth, improve availability when connectivity to the core is intermittent, and keep sensitive data local for privacy or regulatory reasons. The term grew out of late-1990s Content Delivery Network (CDN) work, which pushed static web content into edge caches at points of presence near end users. Through the 2000s and 2010s the same edge footprint was extended to host applications, run shopping carts and ad targeting, and ingest real-time data. In current usage 'edge' covers a spectrum from on-device computation (smartphones, vehicles, industrial sensors) through near-edge micro data centers in cell towers and telco central offices, up to regional CDN PoPs. Intermediate layers between edge and cloud are sometimes called fog computing. Typical applications include Internet of Things workloads where bandwidth or round-trip time to the cloud is prohibitive, autonomous vehicles that must react within tens of milliseconds, real-time video analytics, augmented reality, and on-device machine learning ('edge AI'). On the public internet, CDN operators expose edge compute to developers through serverless platforms such as Cloudflare Workers, Fastly Compute@Edge, and AWS Lambda@Edge, which run short-lived functions in V8 isolates or WebAssembly sandboxes next to the cache, enabling request rewriting, A/B testing, personalization, and authentication without a round trip to a central data center.