Edge Computing

Edge computing places computation and data storage near the source of data, such as users, devices, sensors, or local sites, instead of routing everything to a distant centralized cloud. By shortening the distance data must travel, it reduces latency, saves bandwidth, and enables real-time responses.

How It Works

Workloads run on edge locations: points of presence in many cities, on-premises gateways, telecom base stations, or the devices themselves. A request is handled by the nearest capable node rather than a central region. Many cloud providers offer edge runtimes that execute serverless functions at hundreds of locations worldwide, along with edge caching and storage. Architectures often split responsibilities: time-sensitive or privacy-sensitive processing happens at the edge, while aggregation, long-term storage, and heavy analytics happen in the central cloud. Synchronization and consistency between edge and core are key design concerns.

Why It Matters

Edge computing matters for use cases where milliseconds count or where sending all raw data to the cloud is impractical: real-time personalization, IoT and industrial control, autonomous systems, gaming, video processing, and global APIs that need uniform low latency. It also helps with data-residency requirements by keeping data local. The trade-offs are operational complexity from managing many distributed locations, limited compute at each node, and harder observability. Edge complements rather than replaces centralized cloud.

Related Terms

Edge computing builds on content delivery networks and serverless runtimes and complements centralized regions for latency-sensitive workloads.