What Is IaaS?

How Infrastructure as a Service Works

What is IaaS? It is a cloud model, short for infrastructure as a service, in which a provider rents out basic computing resources: virtual servers, storage and networking. You get these resources on demand and build your own systems on top of them.

The provider owns and operates the physical hardware, the data center, power and cooling, and the virtualization layer that splits large machines into virtual servers. You choose the size of each virtual server, known as an instance type, the operating system image it starts from, the storage attached to it, and how its network is configured. Amazon EC2, Azure Virtual Machines and Google Compute Engine are common examples of this model.

What You Still Manage Yourself

IaaS gives you more control than any other cloud model, and it leaves more responsibility with you. After you launch a virtual server, you are normally responsible for:

• the operating system, including updates and security patches
• middleware such as web servers, database engines and caches
• your applications and their configuration
• backups, encryption and access rules

This split is described by the shared responsibility model. Teams that choose IaaS need system administration skills, or a partner who has them.

Typical Use Cases for Infrastructure as a Service

IaaS fits workloads that need a specific software stack or fine-grained control. Common examples include:

• Legacy applications that depend on a particular operating system version and cannot easily be rewritten.
• Custom web stacks, where a team needs its own web server, database and caching setup.
• Test and development environments that are created for a project and deleted when it ends.
• Peak capacity, where an existing data center runs steady workloads and extra virtual servers are added during busy periods.