What Is Cloud Storage?

How Cloud Storage Works

When you upload a file to cloud storage, it's transmitted over the internet to a data centre, where it's written to disk or flash storage alongside data from many other customers. Providers typically replicate your data across multiple drives, servers, or even entire data centres, so that a single hardware failure doesn't mean lost data.

Most cloud storage is accessed through a web console, a desktop sync client, or directly by applications using an API — commonly the Amazon S3 API, which has become something of an industry standard that many providers support, allowing tools built for one S3-compatible platform to work with another with little or no modification.

Types of Cloud Storage

Benefits of Cloud Storage

• Scalability — Capacity can typically be increased (or decreased) on demand, without having to buy, install and configure new physical hardware.
• Accessibility — Data can generally be accessed from anywhere with an internet connection, supporting remote and distributed teams.
• Reduced capital expenditure — Storage becomes an operating cost rather than a large upfront hardware purchase, since you pay for what you use.
• Built-in redundancy — Reputable providers replicate data across multiple drives or locations, reducing the risk of data loss from a single hardware failure.
• Simplified disaster recovery — Storing a copy of critical data off-site, in the cloud, provides a layer of protection if something happens to your primary location.

Things to Consider

Cloud storage isn't automatically the right answer for every workload, and a few genuine trade-offs are worth weighing up before committing:

• Bandwidth and latency — Uploading and downloading large volumes of data depends on your internet connection. For very large datasets, or applications that need extremely fast, consistent access, on-premises or hybrid storage may perform better.
• Ongoing costs — Cloud storage bills are usually based on the amount stored, plus in many cases charges for data transferred out (egress) and API requests. These can add up over time, and are worth understanding clearly before committing to a provider.
• Data sovereignty and compliance — Where your data is physically stored matters for some regulatory requirements. It's worth confirming exactly which country or region a provider's data centres are located in.
• Vendor lock-in — Some providers make it straightforward to move data elsewhere later; others make it more difficult or costly. S3-compatible platforms tend to make migration between providers considerably easier.

Common Use Cases

Choosing a Cloud Storage Provider

The right provider depends on your specific requirements — capacity, performance, budget, and where your data needs to physically reside. Questions worth asking include: Are there charges for egress or API calls? Is the storage S3-compatible, making future migration easier? Where exactly are the data centres located? What durability and availability guarantees are offered? And is immutable storage available, to protect against ransomware and accidental deletion?