# Cloud Computing

A model for obtaining configurable computing resources over a network on demand with measured, elastic capacity.

- Type: framework
- Domain: Engineering & architecture
- Level: Foundational
- Last reviewed: 2026-09-24
- Original: [Soroush’s Product Wiki](<https://wiki.soroush.uk/topic/cloud-computing/>)

## What it means

Cloud computing is a service model in which shared computing capabilities such as servers, storage, networking, platforms, and software are provisioned over a network with limited manual provider interaction. Common service models include infrastructure, platform, and software as a service; deployment may be public, private, community, or hybrid.

## Why it matters

Product teams can acquire and change capacity faster without owning every physical layer, and can use managed capabilities that would be costly to build. Those benefits exchange capital ownership for recurring cost, provider constraints, shared responsibility, and new skills in governance and operations.

## When to use it

Use cloud services when on-demand provisioning, elasticity, geographic reach, or managed capabilities improve the product and operating model. Dedicated or on-premises infrastructure may fit strict latency, sovereignty, existing investment, disconnected operation, or predictable large-scale economics better.

## How it works

1. Select the service and deployment model that matches control, compliance, and team capability needs.
2. Provision resources through provider interfaces and preferably repeatable configuration.
3. Design identity, networks, data protection, resilience, and observability under shared responsibility.
4. Scale or release capacity as demand changes while tracking service quotas and dependencies.
5. Meter cost and usage by product, environment, or owner and set governance controls.
6. Test portability and recovery assumptions rather than treating provider availability as guaranteed.

## A practical example

A seasonal ticketing product uses managed databases and elastic compute to add capacity before major releases. The team tags resources by product, tests regional recovery, and keeps a cost ceiling because automatic scaling can also amplify inefficient traffic.

## Where it breaks

Cloud services still fail and require skilled operation. Variable bills, data transfer, quotas, jurisdiction, provider-specific services, and migration costs can constrain product choices. Elasticity is not instantaneous or automatically cost-effective.

## Common mistakes

- Treating cloud computing as merely running the same servers in someone else's facility.
- Assuming the provider owns application security, backups, and recovery outcomes.
- Using elasticity without budgets, ownership tags, quotas, or cost monitoring.
- Adopting provider-specific services without evaluating exit and continuity needs.

## Visualization

[View the visualization on the website](<https://wiki.soroush.uk/topic/cloud-computing/#visualization-cloud-computing>)

## Sources

1. [The NIST Definition of Cloud Computing](<https://www.nist.gov/publications/nist-definition-cloud-computing>) — National Institute of Standards and Technology · Government standard
2. [Cloud Computing Synopsis and Recommendations](<https://csrc.nist.gov/pubs/sp/800/146/final>) — National Institute of Standards and Technology · Government guidance

## Related topics

- [Container](<https://wiki.soroush.uk/topic/container/>)
- [Content Delivery Network \(CDN\)](<https://wiki.soroush.uk/topic/content-delivery-network/>)
- [Disaster Recovery](<https://wiki.soroush.uk/topic/disaster-recovery/>)
