Cloud computing explained: types, service models, benefits and real examples

Cloud computing in one paragraph

Cloud computing is the delivery of computing over the internet. Instead of buying a server, finding a room with power and cooling, and replacing parts when they fail, you rent capacity from a provider that runs it at scale. You switch it on in a dashboard, you pay for the period you use it, and you switch it off or resize it when your needs change. Everything from a two-page company website to a machine-learning training run on GPUs can work this way.

The idea is old — shared mainframes did something similar decades ago — but three things made it mainstream: fast, cheap networks; virtualisation, which lets one physical machine safely run many isolated virtual ones; and automation, which lets a provider create a server, a database or a storage bucket in seconds without a person touching hardware.

The five characteristics that make something “cloud”

A useful test comes from the standard definition used across the industry. If a service has these five properties it is cloud computing; if it is missing several, it is probably just hosting with a new label.

• On-demand self-service: you create and change resources yourself, without a sales call.
• Broad network access: you reach it over the internet from any device.
• Resource pooling: the provider serves many customers from shared, isolated infrastructure.
• Rapid elasticity: capacity grows or shrinks quickly, often by changing a plan.
• Measured service: usage is metered and billed, so you pay for what you take.

The four deployment types

Deployment type answers one question: whose infrastructure does it run on, and who else shares it?

Public cloud runs on a provider's shared infrastructure — AWS, Microsoft Azure, Google Cloud, or an Indian provider like Cloud Ninza. It is the cheapest way to start and the fastest to scale. Private cloud is the same technology dedicated to one organisation, either in its own data centre or hosted for it; it suits strict compliance needs and very steady, very large workloads. Hybrid cloud connects the two, typically keeping sensitive systems private while customer-facing apps run in public cloud. Multi-cloud means using more than one public provider, usually to avoid depending on a single vendor or to use one provider's strength — Google Cloud for analytics, say — alongside another's.

For a business website, an online store or a startup's first app, public cloud is almost always the right answer. Private and hybrid start to make sense when a regulator, a contract or sheer scale tells you so.

• Public cloud: shared, pay-as-you-go, fastest to start
• Private cloud: dedicated to one organisation
• Hybrid cloud: public and private connected
• Multi-cloud: more than one public provider

The three service models: IaaS, PaaS and SaaS

Service model answers a different question: how much of the stack do you manage yourself? Think of it as renting a car. IaaS is renting the car and driving it yourself; PaaS is booking a driver who knows the roads; SaaS is taking a train — you only choose where you are going.

Infrastructure as a Service (IaaS) gives you virtual servers, disks and networks. You install the operating system's updates, the web server, the language runtime and your app. You have full control, and full responsibility. A cloud server is IaaS.

Platform as a Service (PaaS) gives you a place to run code. You bring your WordPress site, Node.js app, Django project or Spring Boot JAR; the platform provides the runtime, the web server, SSL, logs, restarts and isolation. Managed cloud hosting is PaaS in practice, and it is where most websites and business apps belong.

Software as a Service (SaaS) is finished software you use in a browser — Gmail, Google Workspace, Microsoft 365, Zoho, Shopify. You manage your data and users, nothing else.

• IaaS: virtual servers — you run everything above the hardware
• PaaS: deploy code — the platform runs it for you
• SaaS: use finished software in a browser

Real benefits — and the honest trade-offs

The benefits are real, but they are not magic, and knowing the trade-offs is what separates a good cloud decision from an expensive one.

• No hardware to buy: costs move from a large upfront purchase to a predictable monthly or yearly bill.
• Speed: a new server, database or site is a few clicks, not a procurement cycle.
• Scale in place: grow memory, storage or plan without moving your site.
• Resilience: providers replace failed hardware underneath you, and backups can live away from the server.
• Security at scale: patching, isolation and monitoring done by specialists — if your provider actually does them.
• Trade-off — cost discipline: pay-as-you-go also means forgotten resources keep billing.
• Trade-off — dependence: your provider's outages and policies become yours, so choose one you can reach.

Common cloud services, with examples

Cloud providers sell dozens of products, but most needs fall into a handful of categories. Compute: virtual servers, GPUs and managed app hosting. Storage: block storage (disks for servers), object storage (files and backups) and archives. Databases: MySQL, MariaDB, PostgreSQL, MongoDB and Redis as managed services. Networking: load balancers, CDNs, DNS and private networks. Plus identity, monitoring, backup and disaster recovery.

A typical small business uses three or four of these without thinking of them as separate products: a managed hosting plan for the website, a database, free SSL, a CDN such as Cloudflare in front, and backups. A growing SaaS company adds cloud servers, object storage and maybe GPUs for AI features.

Which model fits a website, an app or a company?

Match the model to the work, not to the hype. A company website, WordPress blog or WooCommerce store: managed WordPress or PHP hosting (PaaS). A web app or API in Node.js, Python, PHP or Java: managed app hosting with Git deployments (PaaS). Software that needs root access, Docker, always-on background workers or unusual system packages: a cloud server (IaaS). AI training or inference: GPU cloud. Email and office tools: SaaS such as Google Workspace or Microsoft 365.

Most organisations end up with a mix — and that is fine, as long as each piece is chosen on purpose and someone can see the whole bill.

Cloud computing with Cloud Ninza

Cloud Ninza runs managed cloud hosting and cloud servers from India. Managed plans for WordPress, PHP, Node.js, Python and Java cost from ₹1,188 to ₹19,188 a year before GST, each account isolated with its own memory ceiling, with a dashboard that covers files, a browser terminal, databases, Git deployments, logs, cron jobs and free SSL that issues and renews itself. Longer terms cost less: 2 years save 5%, 3 years save 10%, 4 years save 15% and 5 years save 20%.

Cloud servers start at ₹1,073 a month and GPUs at ₹36 an hour, all at published prices you can add up in the cloud price calculator before asking for a quote. Renewals are at the plan's published price on the renewal date, by invoice — nothing is charged automatically.

Key takeaways

The short version, for anyone skimming:

• Cloud computing is renting computing over the internet and paying for what you use.
• Deployment types: public, private, hybrid, multi-cloud. Most businesses start on public cloud.
• Service models: IaaS (servers), PaaS (deploy code), SaaS (finished software).
• Websites and apps usually belong on managed PaaS-style hosting; root-level needs go to a cloud server.
• Choose a provider you can reach, with published prices and backups you can restore.