How Does Cloud Storage Work? A Plain-English Explanation
Cloud storage can seem like magic, but behind the scenes it is a fairly straightforward process.
With cloud storage, your device sends a copy of a file over the internet to servers in the provider’s data centre. There it is usually kept on several drives, and you can retrieve it again with your account. This is how you reach the same files from different devices.
What happens when you upload a file?
Say you take a photo and your cloud storage app is switched on. The app notices there is a new file and starts an upload. The file travels over your Wi-Fi or mobile connection and across the internet to the provider’s servers.
Large files are often split into small pieces. If your connection drops halfway, the app does not have to start all over again. At the other end, the server puts the pieces back together into a single file.
The service then records what has been stored: the name, the size, the time and which account it belongs to. This information is usually kept separate from the content itself. That way the service can quickly show your folders without opening every file.
An everyday example: you keep a PDF of your energy contract in a folder. As soon as you open that folder on another computer, the service asks the server for that file. The server sends it over and you see the document on screen, as if it were on your own drive. In reality it comes from a place that may be hundreds of kilometres away.
Where are your files physically stored?
The cloud is not a cloud at all, but a collection of computers in data centres. These are secured buildings with their own power supply, cooling and fast network connections. A provider uses its own data centres or rents space from another company.
To prevent a faulty drive from causing data loss, many services store data on several drives. Sometimes copies are also kept in several locations. Exactly how this is arranged differs per provider and per plan. It is therefore worth reading the product page and the terms.
The location of the data can also matter to you, for example if your organisation wants to know in which country files are stored. You can read more about this in the guide on data residency.
How do all your devices stay in step?
Much cloud storage works with synchronisation. On your laptop there is a folder linked to your account. If you add a document to it, it is uploaded. Your phone and tablet then get a signal that something new is available and fetch it.
If you change a file, the new version is generally passed on. Many services only send the changed parts, so that it takes less time and uses less data.
Note: synchronisation is not the same as a backup. If you delete something on one device, it may disappear from your other devices too. You can read about the difference in the guide sync or backup.
In the background, the service keeps track of which version of a file is the newest. If you make a change at the same time as another device, a conflict can arise. Many services solve this by keeping two versions, so that nothing is lost. You then see this as a file with something added to its name.
How is your data protected in transit and in storage?
When sending data, services usually use a secure connection (https). This means that someone watching the network cannot simply read the content.
On the servers themselves, files are often stored in encrypted form. Who manages the key, however, differs per service. With some services the provider manages the keys, with others only you do. The latter is called end-to-end or zero-knowledge encryption and is covered in the blog post end-to-end encryption in plain English.
Your own choices also play a part, such as a strong password and two-step verification. These make sure someone else cannot simply get to your files using your account.
One further element is built-in access control. With every request, the server checks who you are and whether you are entitled to the file. That is why you have to log in.
What do you need to use cloud storage?
The basics are simple:
- an account with a provider;
- an internet connection;
- an app, program or web browser to reach your files.
Many providers have a free tier with a limited amount of space and paid tiers with more storage or extra features. Which tier suits you depends on how much you store and what you want to do with it. You can read how to estimate that in how much storage space do I need.
Without internet, you can often only reach files that you have specifically made available offline. It is useful to know this before you travel or sit on a train with poor reception.
What are the main types of cloud storage?
Cloud storage is a broad term. For everyday use, such as photos and documents, a service works with folders and files, much like your own computer. For businesses and developers there is also object storage, in which data is kept as separate objects with a unique name. This is useful for large amounts of data and for applications that retrieve files automatically. You can read more about this in the guide what is cloud storage.
You can also use a provider’s service, or run software yourself on your own hardware. The latter can be done on a NAS at home or in the office, for example. Both approaches follow the same basic principle: data goes over a network to a storage system and comes back when you ask for it. The difference lies in who manages the system.
Frequently asked questions
Is cloud storage the same as a backup?
Not necessarily. Cloud storage keeps files online and often syncs them between devices, whereas a backup is meant to let you recover old or deleted data. Some services offer both, but check what your plan actually does.
Where are my files when I use cloud storage?
They are on servers in the data centres of the provider or of a partner. The location differs per service, and sometimes you can choose a region. The provider’s website or terms explain how this is arranged.
Can I reach my files without internet?
Only if you have made files available offline in the app, or if they are in a locally synced folder. Everything else needs a connection to be retrieved.
Further reading
Last updated on
