Cloud Computing

Cloud computing is the delivery of computing resources, such as servers, storage, databases, networking and software, over the internet on demand, so that customers rent capacity from a provider's data centers instead of buying and running their own hardware. The idea is closely related to older notions of utility computing, but it became mainstream in the mid-2000s when providers exposed their infrastructure through programmable interfaces and billed for what was used by the hour, the gigabyte or the request.

The most cited formal definition comes from the US National Institute of Standards and Technology, whose Special Publication 800-145 (Peter Mell and Timothy Grance, 2011) describes cloud computing by five essential characteristics, three service models and four deployment models. The characteristics are on-demand self-service, broad network access, resource pooling, rapid elasticity and measured service. The deployment models are private, community, public and hybrid clouds. The service models are infrastructure as a service (IaaS), where the customer rents virtual machines, storage and networks and manages the operating system and everything above it; platform as a service (PaaS), where the provider also runs the runtime and the customer only deploys application code; and software as a service (SaaS), where the customer simply uses a finished application, as with webmail or an online office suite.

The commercial turning point was Amazon Web Services, which launched the Simple Storage Service (S3) in March 2006 and the Elastic Compute Cloud (EC2) in beta later that year. S3 offered storage accessed through a web API and charged by volume, and EC2 let anyone rent virtual machines in minutes with a credit card. Microsoft Azure and Google Cloud followed, and the model rested on virtualization, which lets many customers share physical servers safely. The economic shift is from capital expense, buying machines in advance for peak load, to operating expense, paying for capacity as it is used. Elasticity means a service can scale from a handful of machines to thousands and back, which suits traffic spikes and also machine-learning training, where rented GPU clusters are used for a few days and released.

Security in the cloud rests on the shared responsibility model: the provider secures the cloud itself, meaning the physical facilities, hardware and virtualization layer, while the customer remains responsible for what they put in it. The split moves with the service model. In IaaS the customer patches the operating system, configures firewalls and manages access, while in SaaS the provider handles most of the stack but the customer still controls user accounts, sharing settings and data classification. Many well-known data leaks have come from customer misconfiguration, such as a storage bucket accidentally set to public, rather than from a breach of the provider's data center. Identity and access management, encryption and logging are therefore the customer's core duties; the broader discipline of cybersecurity covers them in more depth.

Cloud is not automatically cheaper, more reliable or more private, and treating it as magic is the standard pitfall. Pay-per-use pricing punishes forgotten resources and heavy data transfer, since moving data out of a provider often costs money and makes switching vendors harder, a problem known as lock-in. Reliability is designed rather than given: providers publish availability zones and regions, and applications must be built to survive the loss of one. The cloud is also not an abstract place; it is real buildings full of servers, subject to local law, which matters for data residency and regulation. Modern practice builds on it with containers, serverless functions and managed databases, and most large-scale AI services today are trained and served on rented cloud hardware.

Read cloud computing as renting computing the way households rent electricity: a provider operates the machines and sells measured, elastic slices of them through an API, and the customer's bargain is less hardware to own in exchange for continuing responsibility for configuration, cost and the data placed there.