Cloud Computing
Definition
Cloud Computing
Cloud computing is a model for delivering computing resources on demand over a network, from a shared pool that can be scaled or released without buying hardware. The defining shift is from owning capacity to renting it, and almost every consequence follows from that.
The definition is not loose marketing language. It was standardised early and has held — which is why the vocabulary is consistent across vendors who agree on little else.
Three service models describe how much of the stack the provider operates. Four deployment models describe who the infrastructure is shared with.
The commercial change is as significant as the technical one. Capacity becomes an operating expense that varies with use — which moves cost control from procurement into daily engineering decisions.
Key takeaways
- Cloud computing rents pooled, on-demand computing capacity instead of owning hardware.
- Service models set how much of the stack the provider runs; deployment models set who shares it.
- Cost shifts from capital purchase to variable operating spend tied to consumption.
- Security responsibility is shared, and the dividing line moves with the service model.
How it works
Resources are pooled across many customers and allocated on request, then released when no longer needed. Billing follows measured consumption — so capacity once bought years ahead is now provisioned in minutes and paid for by use.
The reference definition is a standard, not a vendor claim. The National Institute of Standards and Technology published it in Special Publication 800-145, dated September 2011, setting out five characteristics, three service models and four deployment models.
The service models differ in where the provider’s responsibility stops.
| Service model | Provider operates | Customer operates |
|---|---|---|
| Infrastructure as a service | Compute, storage, network | Operating system and everything above |
| Platform as a service | Runtime and middleware too | The application and its data |
| Software as a service | The whole application | Configuration, users and data governance |
Deployment models answer a different question. Public clouds share infrastructure across unrelated tenants, private clouds serve one organisation, community clouds serve a defined group, and hybrid arrangements combine them.
Security is shared, and the split follows the service model.
Federal guidance on the subject runs to a dedicated architecture: the Cloud Security Technical Reference Architecture, version 2, dated June 2022, produced jointly by CISA, the United States Digital Service and FedRAMP.
Cost discipline becomes an engineering habit. Because spending tracks consumption, an unreviewed environment accumulates charges that no procurement cycle ever approved.
Examples
Cloud adoption takes very different shapes depending on what the organisation is optimising for. The cases below show migration, elasticity and regulated workloads side by side.
A retailer moves its storefront to public cloud for elasticity. Peak trading days need many times the baseline capacity, and renting that peak is far cheaper than owning it year-round.
A bank runs a hybrid arrangement. Regulated core systems stay in a private environment while analytics and customer-facing services run in public cloud under defined data boundaries.
A software company builds on platform services from the start. It never operates servers, which removes a whole class of work and replaces it with platform cost management.
A manufacturer lifts legacy applications into cloud infrastructure without redesigning them. Costs often rise initially, because applications built for fixed hardware do not release capacity when idle.
Related terms
Cloud work is described through a set of delivery and management terms that are frequently blurred. The entries below mark which layer and which commercial model each one refers to.
- Cloud Managed Services: the ongoing operation of a cloud environment by an external provider.
- Public Cloud Outsourcing: delivery on shared multi-tenant infrastructure run by a hyperscale provider.
- Cloud Migration Strategy: the plan governing which workloads move, in what order and by what method.
- Cloud Adoption Framework: the structured approach to governance, skills and operating model during adoption.
- Cloud Operating Model: the way teams, responsibilities and controls are arranged once workloads run in cloud.
- Cloud Cost Optimization: the practice of reducing consumption-based spend without losing capability.
FAQ
What are the three cloud service models?
Infrastructure as a service, platform as a service and software as a service. They differ in how much of the technology stack the provider operates on the customer’s behalf.
Is cloud computing cheaper than owning servers?
Not automatically. It converts capital cost into variable operating cost, and an application that cannot scale down often costs more in cloud than it did on owned hardware.
What is the difference between public and private cloud?
Public cloud shares infrastructure across unrelated customers; private cloud dedicates it to one organisation. Hybrid arrangements combine both under defined boundaries.
Who is responsible for cloud security?
Both parties, along a line set by the service model. The provider secures the layers it operates, and the customer secures configuration, access and data.
Does moving to cloud remove the need for IT staff?
No, it changes the work from hardware operation to platform engineering, security and cost management.
What makes a migration go over budget?
Moving applications unchanged.
Compare cloud and technology providers through Outsource Accelerator to see which delivery models fit the workload.
Independent
