They are critical components of how applications run, dictating the level of computing power, storage, and memory they have access to. This makes them central to preparing for the AWS Certified Solutions Architect – Associate (SAA-C03) exam.

Table of Contents

Instance Types

Instance types determine the hardware of the host computer used for the instance. AWS provides a variety of instance types optimized to fit different use cases. They balance the suite of compute, memory, storage, and networking resources to provide optimum performance for different workload requirements.

Instance Families

On the other hand, instance families are categorized based on targeted workload solutions. Here are some of the instance families that AWS provide:

  • Compute Optimized: The C4, C5, C5n, C6g, C6gn, and C6i instance families are geared towards compute-intensive workloads like high performance web servers, batch processing, ad serving, video encoding, distributed analytics, high-performance computing (HPC), and machine learning inference.
  • Memory Optimized: R4, R5, R5a, R5b, R6g, and R6i instance families fulfill the needs of memory-intensive applications like high-performance databases, data mining & analysis, in-memory caches, and real-time big data analytics.
  • Storage Optimized: I3, I3en, D2, and H1 instances offer direct-attached storage and are ideal for big data analytics, NoSQL databases, and data warehousing.
  • GPU Instances: Provide you with access to NVIDIA Tesla GPUs and are intended for GPU-based workloads such as graphics processing or data analysis.

Instance Sizes

Instance sizes, meanwhile, refer to the proportion of CPU, memory, storage, and network resources allocated to a particular type of instance. They usually follow a particular scheme:

  • Standard (m5): Balanced CPU-to-memory ratio. Ideal for web servers, dev/test servers, data logging, and warm standby databases.
  • Large (m5.large): 2 vCPU and 8 GiB memory.
  • Extra Large (m5.xlarge): 4 vCPU and 16 GiB memory.

The naming convention is such that larger instance sizes within an instance family offer more of these computational resources.

Virtualization Types

There are two types of virtualization: paravirtual (PV) and hardware virtual machine (HVM). HVM instances are designed to take advantage of hardware virtualization extensions, providing the ability to run with better performance when running operations that require full virtualization.


As you can see, understanding AWS instance types, families, and sizes is essential for any solutions architect. They form the backbone of AWS services and determine how effective your applications are on the cloud. With a clear understanding of the resources available to you, you can more effectively design, manage, and optimize your architecture. Always remember that choosing the right instance is paramount to striking a balance between cost efficiency and a seamless, powerful application performance.

Practice Test

True or False: The t2 instance type on AWS is memory-optimized.

  • True
  • False

Answer: False

Explanation: The t2 instance type on AWS is not memory-optimized, rather it is a general-purpose instance type and is suitable for small databases and development environments.

Which AWS instance family is compute-optimized?

  • a. M4
  • b. T2
  • c. C4
  • d. R4

Answer: c. C4

Explanation: The C4 instance family is designed for tasks requiring high computation capabilities like high-performance web servers and batch processing.

True or False: D3 instance types are designed for workloads that require higher memory.

  • True
  • False

Answer: False

Explanation: D3 instance types are not designed for higher memory workloads. Rather, they are designed for workloads that require high disk throughput and I/O.

Which of these options is NOT a valid Amazon EC2 instance type?

  • a. x1
  • b. d2
  • c. r4
  • d. y2

Answer: d. y2

Explanation: Amazon EC2 does not have a y2 instance type.

Using a smaller instance size incurs more cost as compared to a larger one. True or False?

  • True
  • False

Answer: False

Explanation: Typically, smaller instances are cheaper as they provide fewer resources.

True or False: In terms of virtualization in AWS, there are two types: hardware virtual machine (HVM) and paravirtualization (PV).

  • True
  • False

Answer: True

Explanation: Yes, AWS offers two types of virtualization: HVM and PV. HVM virtualization fully emulates the hardware, while PV involves time-sharing system resources.

Amazon EC2 R5 instances are designed for which of the following?

  • a. High-performance computing
  • b. Memory-intensive applications
  • c. General purposes
  • d. High storage needs

Answer: b. Memory-intensive applications

Explanation: R5 instances are memory-optimized and suit applications needing large amounts of RAM, like high-performance databases and in-memory analytics systems.

True or False: Each instance type offers a fixed combination of CPU, memory, storage, and network capacity.

  • True
  • False

Answer: True

Explanation: Each instance type in AWS offers different combinations of these resources to suit diverse needs, and they are fixed per type.

A GPU instance type is best utilized for:

  • a. Data caching
  • b. Web serving
  • c. Machine learning
  • d. Storage databases

Answer: c. Machine learning

Explanation: GPU (Graphics Processing Unit) instances are best used for graphic-intensive tasks, including machine learning, 3D rendering, and game streaming.

AWS virtual machines are called instances. True or False?

  • True
  • False

Answer: True

Explanation: In AWS, the virtual servers rented by customers are termed ‘instances’.

Interview Questions

What are instance types in AWS?

Instance types in AWS determine the hardware of the host computer used for the instance. Each instance type offers different compute, memory, and storage capabilities and are classified into different families based on these capabilities.

What are AWS Instance Families?

AWS Instance Families are groupings of instance types optimized to fit different types of applications and use cases. They include General Purpose, Compute Optimized, Memory Optimized, Storage Optimized, and Accelerated Computing instances.

What is a memory-optimized instance in AWS?

Memory-optimized instances in AWS are designed to deliver fast performance for workloads that process large data sets in memory. They are ideal for relational databases, in-memory databases, and other enterprise-level applications with high memory requirements.

Can you name a few AWS Compute Optimized instances and their ideal use cases?

AWS Compute Optimized instances are C3, C4, C5, and CC2. They are ideal for compute-bound applications that benefit from high-performance processors, such as batch processing workloads, media transcoding, distributed analytics, high-performance web servers, high-performance computing (HPC), scientific modeling, dedicated gaming servers, and ad server engines.

Define virtualization in the context of AWS?

Virtualization in AWS is the capability that enables you to create and manage Virtual Private Clouds (VPCs) and EC2 instances. Amazon EC2 provides two types of virtualization – paravirtual (PV) and hardware virtual machine (HVM).

How does AWS classify instances based on size?

AWS classifies instances based on size to offer scalability according to the workload. The instances range from nano instances (smallest) to large instances, each offering different levels of compute, memory, and storage suited for different workloads.

What are accelerated computing instances in AWS?

Accelerated computing instances in AWS utilize hardware accelerators, or co-processors, to perform functions, like graphics processing or data pattern matching, more efficiently than is possible in software running on CPUs.

What are storage optimized instances in AWS?

Storage optimized instances in AWS are designed for workloads that require high, sequential read and write access to large data sets on local storage. They are optimized to deliver tens of thousands of low-latency, random I/O operations per second (IOPS) to applications.

What are general-purpose instances in AWS?

General-purpose instances in AWS provide a balance of compute, memory, and network resources and are a good choice for many different types of applications such as web servers, microservices, and backend servers for enterprise applications.

What are burstable performance instances in AWS?

Burstable performance instances in AWS are designed to provide a baseline level of CPU performance with the ability to burst to a higher level when required by your workload. They are suited to a wide range of general-purpose applications.

Which type of virtualization should be used for Windows Instances on AWS, PV or HVM?

For Windows instances on AWS, HVM (Hardware Virtual Machine) is the recommended choice of virtualization.

What is the primary advantage of using a compute-optimized instance in AWS?

The primary advantage of using a compute-optimized instance in AWS is its ability to provide high-performance processors. These are suitable for compute-bound applications that require high-performance front-end fleets, web servers, batch processing, distributed analytics, high-performance science and engineering applications, ad serving, MMO gaming and video encoding.

What kind of workloads are suitable for memory-optimized instances?

Memory-optimized instances are suitable for workloads with high memory requirements like in-memory databases, data mining & analysis, caching fleets, and real-time big data analytics.

How are storage-optimized instances typically used?

Storage-optimized instances are typically used in workloads that require high, sequential read and write access to very large data sets on local storage. They are ideal for distributed file systems, data warehousing, and high-frequency online transaction processing (OLTP) systems.

What is the benefit of using accelerated computing instances?

Accelerated computing instances are useful for applications requiring parallel computing capabilities. They provide access to NVIDIA GPUs to deliver high-performance graphics and processing capabilities, making them ideal for machine learning, computational fluid dynamics, computational finance, seismic analysis, rendering, molecular modeling, genomics, and other server-side GPU compute workloads.

Leave a Reply

Your email address will not be published. Required fields are marked *