What is Cloud Computing?

Cloud computing is an abstraction of compute, storage, and network infrastructure assembled as a platform on which applications and systems can be deployed quickly and scaled on the fly. Crucial to cloud computing is self-service: Users can simply fill in a web form and get up and running.

The vast majority of cloud customers consume public cloud computing services over the internet, which are hosted in large, remote data centers maintained by cloud providers. The most common type of cloud computing, SaaS (software as service), delivers prebuilt applications to the browsers of customers who pay per seat or by usage, exemplified by such popular apps as Salesforce, Google Docs, or Microsoft Teams. Next in line is IaaS (infrastructure as a service), which offers vast, virtualized compute, storage, and network infrastructure upon which customers build their own applications, often with the aid of providers’ API-accessible services.

When people casually say “the cloud,” they most often mean the big IaaS providers: AWS (Amazon Web Services), Google Cloud, or Microsoft Azure. All three have become gargantuan ecosystems of services that go way beyond infrastructure: developer tools, serverless computing, machine learning services and APIs, data warehouses, and thousands of other services. With both SaaS and IaaS, a key benefit is agility. Customers gain new capabilities almost instantly without capital investment in hardware or software—and they can instantly scale the cloud resources they consume up or down as needed.

Cloud computing definitions for each type

Way back in 2011, NIST posted a PDF that divided cloud computing into three “service models”—SaaS, IaaS, and PaaS (platform as a service)—the latter a controlled environment within which customers develop and run applications. These three categories have largely stood the test of time, although most PaaS solutions now make themselves available as services within IaaS ecosystems rather than presenting themselves as their own clouds.