Amazon Web Service (AWS) has seen significant growth since its inception. It is one of the popular cloud computing services used by organisations and developers. According to reports, during the third quarter of last year, the revenues of the cloud service grew 35% compared to the preceding quarter.

As of today, most of the industry works have been spreading over the cloud network, and organisations are spending heavily on computations and storage in the cloud. It thus becomes essential that one must think about the optimisation of cost while using cloud services. 

Cost optimisation is a technique of achieving an output at the lowest cost possible while utilising all the required resources. To use a cost-optimised cloud, one must use the appropriate and specific services and resources, as well as configurations of a particular workload. In the cloud, the cost optimisation is composed of mainly four areas, which are

• Cost-effective resources
• Matching supply with demand
• Expenditure awareness
• Optimising over time

In this article, we list down a few essential points which will help you in optimising the cost of AWS cloud service.

The list is in no particular order

1| Attribute Provisioning

When provisioning systems on cloud services, one requires a proper understanding of the requirements of adjusting the service capacity. In AWS, users can modify the attributes of AWS managed services using the AWS Management Console or AWS APIs as well as SDKs to align resource according to the changing demand. They can also pack multiple instances on an AWS resource to enable higher density usage. Amazon CloudWatch is a vital AWS service which allows a user to collect as well as track metrics. 

2| Right-Sizing

Right-sizing helps to meet the service with the exact capacity requirements without compromising the capacity. Right-sizing means to use the lowest cost resource, which meets the technical specifications of a specific workload. 

While performing right-sizing exercises, one must consider that the monitor must accurately reflect the end-user experience, assess the cost of modification against the benefit of right-sizing and select the correct granularity for the period of analysis which is required to cover any system cycles. AWS provides several instance-based services which allow resources to be modified as the demands change. For instance, Amazon Elastic Compute Cloud (Amazon EC2), Amazon Elastic Block Store (Amazon EBS), and RDS. 

3| On-Demand Instances

With on-demand cases, one can increase and decrease the capacity of the resources one requires based on the demand of the application. On-demand is recommended for applications which include short-term workloads and are also suitable for workloads related to the development and testing environments.  

4| Spot Instances

In AWS, Spot instances allow a user to use the spare compute capacity at a significantly lower cost. While using Spot instances, one must consider the instance type, the location of the resource and the design for continuity.

It can be launched with some required time duration and are ideal for use cases such as batch processing, scientific research, image or video processing, financial analysis, and testing. Spot instance can also be used to reduce the cost of running applications by increasing the computing scale and throughput for the particular budget.  

5| Reserved Instances

Reserved instance is one of the ways to reduce the costs by reserving a specific amount of resources in the cloud platform. With this method, a user can commit to a certain period of usage and can save the hourly rate expenses. For instance, there are two types of Amazon EC2 Reserved instances which are

• Standard: Standard Reserved Instances provide both a billing benefit and capacity reservation when the instance family, size, and availability zone are specified.
• Convertible: Convertible Reserved Instances are provided for a one-year or three-year term and allow conversion to different families, new pricing, different instance sizes, different platforms, or tenancy during the period.  

6| Monitoring and Managed Services

AWS managed services operate at cloud scale and offer a lower cost per transaction or service in the cloud. There are various managed services for databases in this platform, such as Amazon RDS and Amazon DynamoDB. There are also serverless or application-level services such as AWS Lambda, Amazon Simple Queue Service (Amazon SQS), Amazon Simple Notification Service (Amazon SNS), and Amazon Simple Email Service (Amazon SES), which help in eliminating the need for managing virtual servers for code execution, queuing services and message delivery.   

7| Optimise Data Transfer

To minimise the cost of data transfer, one must define the architecture of data transfer. This can be done by using Amazon CloudFront, which is a global content delivery network that delivers data with low latency and high transfer speeds. 

8| Turn It Off Or Schedule It

This one of the most critical points to have a cost-optimised workload. One can shut down the unused instances while working in developing environments. One can also opt for scheduling on and off times when working on non-production cases such as testing, developing, etc. Following this will save up to 70% of the costs in the cloud.