The cloud is someone else’s computers
Despite the name, the cloud is not a vague presence in the sky. It is a large collection of real computers, storage devices and network equipment, housed in buildings and run by a company that rents access to them over the internet. When you use a web email service, stream a video or save photos to an online account, your device is talking to machines somewhere else and showing you the result.
Cloud computing is the practice of using those remote machines, instead of computers you own and maintain yourself, to store data, run software and perform calculations. The key shift is from owning equipment to renting capability.
A simple comparison
Think about electricity. A factory could build its own power plant, or it could connect to the grid and pay for what it uses. Most do the latter. Cloud computing applies the same idea to computing: rather than buying servers, installing them in a back room and replacing them every few years, an organization uses a provider’s equipment and pays for the portion it needs.
The main building blocks
Most cloud services are built from a handful of ingredients:
- Compute: processing power that runs programs, from a small website to a large calculation.
- Storage: places to keep files, databases and backups.
- Networking: the connections that link those machines to each other and to users.
- Management tools: dashboards and programming interfaces that let customers start, stop and configure resources.
All of this lives in facilities that are often called data centers, which are buildings full of racks of servers with power, cooling and security systems.
How sharing works: virtualization
A single physical server is powerful, and most workloads do not use all of it. Providers solve this with virtualization, software that divides one physical machine into several isolated virtual machines. Each virtual machine behaves like a separate computer with its own operating system, even though they share the same hardware underneath.
Many customers can therefore share the same pool of equipment, each seeing only their own slice. This sharing is often called multi-tenancy. Providers add isolation controls so one customer cannot see another’s data, though how well any particular setup is configured is still something customers and providers need to get right.
Newer approaches such as containers package software more lightly than full virtual machines, but the principle is similar: use shared hardware efficiently.
Characteristics that define the cloud
Widely used definitions describe a few traits that distinguish cloud computing from simply hosting a server somewhere:
- On-demand self-service: a customer can set up resources without phoning a salesperson or waiting for equipment to be delivered.
- Broad network access: services are reached over the internet from many kinds of devices.
- Resource pooling: the provider serves many customers from a shared pool.
- Rapid elasticity: capacity can be increased or reduced quickly as needs change.
- Measured service: usage is tracked, which is what makes billing by use possible.
Why organizations use it
Organizations often turn to cloud services for a mix of reasons. They can start small without a large upfront purchase, add capacity during busy periods, and reach users in different regions. Teams can also hand off some routine work, such as replacing failed hardware, to the provider. Many companies find this lets staff spend more time on their own products rather than on equipment.
Cloud platforms are also commonly used for workloads such as analysis of big data, running machine learning experiments, and supporting remote work tools, because those tasks can need a lot of capacity for short periods.
What organizations weigh on the other side
The cloud is not automatically cheaper or safer, and it is not right for every situation. Common concerns include:
- Cost behavior: paying for usage can be flexible, but bills can grow in ways teams do not expect if resources are left running.
- Dependence on a provider: moving systems elsewhere later can be difficult.
- Security responsibilities: the provider secures its facilities and core platform, but customers remain responsible for how they configure and use the services.
- Connectivity: cloud services depend on a working internet connection, and delays can matter for some uses.
- Location of data: laws and policies about where information may be stored can affect choices, which is the concern behind data sovereignty.
- Environmental impact: data centers use significant energy, and providers and customers both influence the overall digital carbon footprint.
None of these are reasons to avoid the cloud. They are the questions a careful organization asks before relying on it.
Cloud is a spectrum, not a single product
People use the word “cloud” for many different things: a place to store holiday photos, a platform that runs a bank’s applications, or a complete business program accessed through a web browser. These differ mainly in how much the customer manages and how much the provider handles. Services can also be offered publicly to many customers or reserved for a single organization, and some organizations combine several approaches.
Those distinctions, along with questions about cost, security, moving existing systems and avoiding being tied to one supplier, are what the rest of this topic area explores. A good starting point is always the same: know what you are renting, who is responsible for each part, and what it would take to change course.
A note on advice
This article is general information. Decisions about cloud services involve legal, security, financial and technical considerations that differ by place and organization, so it is sensible to consult qualified professionals before making them.