What is an Instance in Cloud Computing?

An instance in cloud computing is a server resource provided by third-party cloud services. While you can manage and maintain physical server resources on premises, it is costly and inefficient to do so. Cloud providers maintain hardware in their data centers and give you virtual access to compute resources in the form of an instance. You can use the cloud instance for running compute-intensive workloads like containers, databases, microservices, and virtual machines.

Why are cloud instances important?

A cloud instance allows software developers to scale beyond traditional physical boundaries. Unlike physical servers, developers don’t need to worry about the underlying hardware when deploying workloads on a cloud instance. There are two main benefits of cloud instances.


Developers scale computing resources in a cloud instance according to their workload requirements. For example, software developers deploy an application on an instance. As the app gains more users, it experiences huge traffic that slows down response time. Developers can horizontally scale cloud resources by increasing the CPU, memory, storage, and network resources to the particular instance. 

Fault tolerance

Organizations create redundancy by using multiple duplicate instances for backup. They are especially useful for managing memory-intensive workloads like data processing. For example, an application can still run on other instances in the US and Asia if a cloud instance hosted in Europe fails.  

What types of workloads can you run on a cloud instance?

Compute intensive

You can run high performance computing workloads on instances, such as distributed analytics, machine learning (ML) algorithms, batch processing, ad serving, video encoding, scientific modeling, and scalable multi-player gaming applications.

Memory intensive

Instances are useful for running memory-intensive workloads such as real-time data ingestion, distributed in-memory caches, big data analytics, memory-intensive enterprise applications, and high-performance databases.

Graphics intensive

Applications that render graphics require high processing and storage capabilities. You can run virtual reality applications, 3D rendering, animation, computer vision, video streaming, and other graphics workloads on a cloud instance.

How do cloud instances work?

A cloud instance abstracts physical computing infrastructure using virtual machine technology. It is similar to having your own server machine in the cloud. You basically create and manage your own virtual server instance in the cloud computing environment. You can configure this cloud server to meet your memory, graphics processing, CPU, and other requirements.

The steps for creating a new instance are:

  1. You use a visual interface or API calls to programmatically create instances
  2. You specify the resources you require or use pre-existing instance types that your cloud provider defines
  3. You can then host your own operating system and other software applications on an instance

The cloud provider will typically charge you only for the resources you actually use. You can create and destroy as many instances as you like. For example, you can use Amazon Machine Images (AMI) to configure and launch cloud instances on AWS. 

What is the instance life cycle?

Developers use a series of steps to set up, run, manage, and stop an instance. The following stages describe an instance life cycle.


Provisioning an instance means setting the compute resources that the instance requires. When developers launch a provisioned instance, it goes into a pending stage. 


At this stage, the instance is deployed and active on the cloud. Developers can deploy workloads such as containerized applications on running instances. They are billed the moment an instance starts running.


Developers might stop an instance to troubleshoot issues that affect the workloads that run on it. When they stop an instance, it enters the stopping stage before being completely halted. Developers can modify the setting of the instance setting when it’s stopped.


Developers can shut down an instance when it is no longer in use. By shutting down an instance, the cloud platform prepares to terminate the instance and remove its corresponding data in the instance store volume. The instance store volume is temporary storage that resides on the same computer as the instance.

What is a preemptible instance?

A preemptible instance is a virtual server that uses spare and nonreserved compute resources. Spare and nonreserved computer resources allow the preemptible instance to be terminated after a specific duration or if the cloud infrastructure reassigns the allocated resources. Developers use preemptible instances for noncritical tasks, such as batch processing, data analysis, or testing.

What is an instance group?

An instance group is a collection of many instances that share the same configuration. Developers use instance groups to set uniform policies and rules across multiple instances easily. All instances in an instance group go through the same lifecycle simultaneously. 

Single instance vs. multi-instance 

Single instance describes a cloud infrastructure setup where only one instance of the compute resources is created and allocated to the user. Meanwhile, multi-instance is a term used for more than one identical instance that run parallel to each other. In a multi-instance setup, each instance has its own compute resources. 

Single tenant vs. multi-tenant

Single tenant means the cloud infrastructure is exclusive to one user. This is common on an on-premises server, where the organization might run multiple workloads on a single machine. Meanwhile, a multi-tenant setup shares the cloud infrastructure amongst multiple users. This allows them to scale workloads on the cloud at a lower cost. 

What is a virtual machine?

A virtual machine is a computer that exists in the virtual environment. Virtual machines use the compute resources that are available in the physical computer. However, those compute resources, such as processing power and memory, are not in the form of physical hardware. Instead, they are provided and managed as code within the physical computer. 

Instance vs. virtual machine

You can run multiple virtual machines on a single computer, but when you run virtual machines in the cloud environment, they are known as instances. Running virtual machines on the cloud allows organizations to benefit from the cost effectiveness of sharing and scaling resources. 

What are the instance types on AWS?

AWS Elastic Compute Cloud (Amazon EC2) allows developers to build scalable apps in the cloud environment. Amazon EC2 offers several types of instances that are optimized for different workloads. 

General purpose instance

A general purpose instance type is ideal for applications that require an equal share of processing power, memory, and networking resources. For example, organizations use a general purpose instance for web servers.

Compute optimized instance

Compute optimized instances are suitable for processing-intensive applications that require high computing power. For example, developers deploy gaming servers, science modeling, and media transcoding workloads on compute optimized instances. 

Accelerated computing instance

Accelerated computing instances provide hardware accelerators, and are suitable for graphics-intensive or analytical workloads. For example, developers build and deploy machine learning (ML) applications on an accelerated computing instance. 

Memory optimized instance

Memory optimized instances allow developers to deploy memory-intensive applications on AWS without compromising performance. For example, developers use memory optimized instances for big data analytics. 

Storage optimized instance

Storage optimized instances include persistent storage to write and retrieve information from databases at a very high speed. For example, developers build search engine functions with a storage optimized instance.

Get started with instances on Amazon EC2 by creating a free AWS account today.

AWS Cloud Instances next steps

Check out additional product-related resources
View free offers for Compute services in the cloud 
Sign up for a free account

Instantly get access to the AWS free tier. 

Sign up 
Start building in the console

Get started building in the AWS Management Console.

Sign in