How does cloud storage work?

Cloud storage provides a modern solution for storing and accessing photos, videos, or documents outside of your local system—without requiring your own hardware. But how exactly do services like this work? And what are the pros and cons of storing data in the cloud?

What is cloud storage?

Cloud storage refers to storage space provided to users via the internet. Unlike local storage on your device, this storage is hosted on external servers, typically managed by spe­cial­ized providers. These providers offer storage space either for free or at a monthly or annual cost. This service model, where the scope and cost of storage can be adjusted as needed, is known as public cloud storage.

An al­ter­na­tive is private cloud storage, which operates on servers within a private network—usually a company network—and is ac­ces­si­ble only to users within that network. In this case, the operator (the company itself or a third party) maintains full control over the storage but is also re­spon­si­ble for the necessary hardware and ad­min­is­tra­tion. A third option, hybrid cloud storage, combines internal cloud storage with external online storage. This is es­pe­cial­ly appealing for busi­ness­es needing internal storage for sensitive data alongside publicly ac­ces­si­ble, easily scalable ca­pac­i­ties.

How does cloud storage work?

Re­gard­less of the type, the basic principle of cloud storage is as follows: The storage provider offers an IT in­fra­struc­ture that ensures smooth and secure server man­age­ment. These servers operate col­lec­tive­ly rather than as stand­alone systems. To achieve this, storage (along with other com­po­nents like memory and CPU) is vir­tu­al­ized using hy­per­vi­sors. These ab­strac­tion layers, also called virtual machine monitors (VMMs), bridge the gap between physical hardware and virtual en­vi­ron­ments. Hy­per­vi­sors are divided into two types, type-1 and type-2 hy­per­vi­sors:

• Type-1 hy­per­vi­sors (bare-metal): Vir­tu­al­iza­tion software that runs directly on the host system without requiring a pre-installed operating system. Examples include VMware ESXi.
• Type-2 hy­per­vi­sors (hosted): Vir­tu­al­iza­tion software that requires a fully installed operating system on the host. It uses the OS’s device drivers for hardware access. Examples include Oracle VM Vir­tu­al­Box.

Accessing vir­tu­al­ized storage typically involves dedicated software. Public cloud storage services usually offer not only web ap­pli­ca­tions ac­ces­si­ble via a browser but also platform-specific apps for various devices. These allow users to log in and access their storage. Files already saved can be retrieved from any supported device (computer, tablet, smart­phone, etc.) as long as there is an active internet con­nec­tion. In private cloud storage, access generally requires a con­nec­tion to the re­spec­tive intranet or VPN (Virtual Private Network).

What is the cloud storage ar­chi­tec­ture?

Given the high demands cloud storage solutions must meet, providers rely on storage ar­chi­tec­tures that out­per­form tra­di­tion­al file-based systems when it comes to ef­fi­cien­cy. While standard file systems work well for basic tasks, they struggle to handle the man­age­ment of billions of files with as­so­ci­at­ed metadata. Cloud storage services thus utilize two al­ter­na­tive formats:

1. Block storage (block-based storage)
2. Object storage (object-based storage)

Block storage, a struc­tured data solution

Block storage is a flexible and logically struc­tured storage model in which all data is divided into uniformly sized data blocks, each assigned its own address. The addresses of the physical storage cells are ab­stract­ed, making the physical location of the storage units ir­rel­e­vant for data storage. The specific hard drive or server where new data is stored, how in­di­vid­ual blocks are arranged, and how access to these blocks is managed can all be easily con­trolled through software. In the cloud, block storage is par­tic­u­lar­ly well-suited as a storage format for database ap­pli­ca­tions and other programs that process struc­tured data.

Object storage, or­ga­niz­ing un­struc­tured data

Object storage is a storage system that saves files as objects along with their cor­re­spond­ing metadata. Each object is assigned a unique iden­ti­fi­ca­tion number, allowing it to be accessed by ap­pli­ca­tions without requiring ad­di­tion­al access control. Stored objects cannot be modified—any changes result in the creation of a new object con­tain­ing the updated version of the original file. This makes object storage an ideal cloud storage format for online backup and archival solutions. Ad­di­tion­al­ly, object-based man­age­ment is well-suited for storing mul­ti­me­dia files like music, videos, or movies that are primarily read-only (e.g., when using streaming services).

Ad­van­tages of cloud storage

Storing files in the cloud can be ad­van­ta­geous for many reasons. One of the most sig­nif­i­cant benefits of this storage solution is that it elim­i­nates the need to purchase and manage dedicated hardware. Unless you opt for a private cloud solution, the provider is re­spon­si­ble for all un­der­ly­ing in­fra­struc­ture. Using cloud storage for data backups offers an ad­di­tion­al advantage: original files and backups are stored in separate locations, reducing the risk of complete data loss due to fire or theft. Other key ad­van­tages of cloud storage include:

• Flex­i­bil­i­ty: You can rent cloud storage as needed and cancel it when it’s no longer required. In contrast, owning hardware often leaves you with unused resources when your storage needs decrease.
• Scal­a­bil­i­ty: Vir­tu­al­ized storage en­vi­ron­ments allow for precise ad­just­ments to storage capacity, which can be scaled up or down depending on your re­quire­ments.
• Avail­abil­i­ty: Public cloud storage is ac­ces­si­ble anytime, from any device, as long as you have an active internet con­nec­tion. This means you can even access your files while on the go.

Dis­ad­van­tages of cloud storage

Cloud storage is a com­pelling and powerful al­ter­na­tive to tra­di­tion­al storage solutions in many ways. However, there are scenarios where storing data in the cloud comes with dis­ad­van­tages. The main dis­ad­van­tage is de­pen­dence on an internet con­nec­tion or, in the case of private cloud storage, the intranet con­nec­tion. Stability is a key factor—when the con­nec­tion fails, access to cloud storage is also in­ter­rupt­ed. Ad­di­tion­al­ly, the available bandwidth plays a crucial role in ensuring smooth data trans­mis­sion, which can be par­tic­u­lar­ly chal­leng­ing with mobile network access. Other dis­ad­van­tages of cloud storage include:

• Provider de­pen­dence: Choosing an external cloud storage solution means relying on the re­spec­tive provider. Unwanted changes to the service offering or even dis­con­tin­u­a­tion of the service cannot be ruled out.
• Security: Trans­mit­ting data across network bound­aries in­tro­duces security risks—not all providers offer TLS/SSL en­cryp­tion for stored data. Ad­di­tion­al­ly, the provider’s IT in­fra­struc­ture can become a high-profile target for cy­ber­at­tacks.
• Privacy concerns: When it comes to data privacy, cloud storage poses inherent chal­lenges. Many providers’ servers are located outside of certain ju­ris­dic­tions, and storing user data may not always comply with local legal re­quire­ments.

Leading cloud storage providers