I recently started teaching a new AWS Cloud Practitioner cohort for NCUS-TEC using resources from AWS Academy. This week we discussed some of the strategies available for cost optimization. During this module, I introduce the DAO Research Whitepaper from 2019 titled Amazon Web Services Cost Surprises Unexpected Costs Experienced by AWS Enterprise Customers. Understanding the pricing model, monitoring usage, and optimizing resources are crucial for effective cost management. There are three lessons that can be learned and gathered from the lesson and whitepaper about the importance of understanding how to navigate the complexities of AWS cost management:
- Understanding the pricing model
- Monitor your usage
- Optimize your resources
1. Understand the Pricing Model
AWS offers various pricing options, including on-demand, reserved instances, and spot instances. It’s essential to learn about these options and how they apply to your specific use case. By understanding the pricing model, you can choose the most cost-effective option for your needs. Accessing your AWS cost and usage information and organizing your AWS implementation for cost management are essential steps in this process
AWS Cost and Usage Report
The AWS Cost and Usage Reports (AWS CUR) provides a fine-grained review of the costs being created within an AWS account. The report is sent to an Amazon S3 bucket and can provide a detailed breakdown based on hour, day, or month. It further segments costs based on the product, product resource, or tags.
2. Monitor Your Usage
Keeping track of your usage is vital for effective cost management. AWS provides tools such as Amazon CloudWatch and AWS Trusted Advisor to help you monitor your usage. By monitoring your usage, you can identify areas where you may be overspending and take steps to reduce your costs. Setting budgets and being alerted when usage is out of bounds using AWS Budgets is also a great way to monitor your costs.
AWS Trusted Advisor
AWS Trusted Advisor takes advantage of best practices that have been collected from serving hundreds of thousands of AWS customers. Trusted Advisor inspects your AWS environment, and then makes recommendations based on observations of the environment and from the machine-generated checks on the AWS Services used. It provides insights on areas of improvement in six different categories.
- Cost optimization
- Performance
- Security
- Fault tolerance
- Service limits
- Operational excellence
While Trusted Advisor is available automatically when you create an AWS account, the level of details and suggestions are based on the support plan chosen. To receive recommendations based on the full set of checks, a Business support plan or higher is required.
There are over 20 cost optimization metrics that Trusted Advisor can advise on. A simple example of these metrics is Amazon EBS over-provisioned volumes. Amazon Elastic Block Store (EBS) is AWS’s block storage service that requires capacity to be provisioned based on anticipated demand. Costs are based on capacity provisioned – regardless if utilized or not. Checks made on the EBS volumes alerts you if any EBS volumes were over-provisioned for your workloads. When you have over-provisioned volumes, you’re paying for unused resources.
3. Optimize Your Resources
AWS offers a wide range of services and features, and it’s important to choose the right ones for your needs. By optimizing your resources, you can reduce your costs while still meeting your performance requirements. For example, using auto-scaling to adjust your resources based on demand or utilizing Amazon S3 lifecycle policies to move data to lower-cost storage tiers can help optimize your resources and reduce costs.
One key point, I have my class look at is evaluating EC2 instance performance. The whitepaper referenced a health services company that stated the need to overprovision their AWS environments, relative to their existing on-premises datacenters. This was to achieve the same performance when moving workloads to AWS and not having compared the performance until after the fact. The performance impact was based not only on compute and storage but often on network bottlenecks. Many customers found that their AWS infrastructure costs can actually increase by 2× their expectation to compensate for the lower performance of AWS versus their existing on-premises environment.
When we begin learning about compute services like EC2, we look at the different types of instances available and their use cases. While the whitepaper, does not provide specific details on the environment and use cases it it a good time to discuss how strategy and analyzing different instance types would best suit their needs.
Consider this…an organization needs resources to support AI applications. This requires computing power to handle complex computations quickly. Using an accelerated computing instance type instead of a general-purpose one can save money by providing more efficient performance for specific workloads. Accelerated computing instances use hardware accelerators, such as GPUs which are better suited for AI and machine learning due to their specialized architecture. This allows them to handle complex computations more efficiently than CPUs.
In conclusion, understanding the pricing model, monitoring your usage, and optimizing your resources are essential components of effective AWS cost management. By implementing these strategies, organizations can develop a better understanding of how to navigate the complexities of AWS cost management and make informed decisions to optimize costs while leveraging the benefits of cloud services.
Vocabulary
AWS and technology is full of acronyms and terminologies that you may not be familiar with. Below are some definitions to help.
- Reserved Instances: Billing discount applied to the use of On-Demand Instances in your account. Learn more at https://docs.aws.amazon.com/whitepapers/latest/cost-optimization-reservation-models/amazon-ec2-reserved-instances.html
- Spot Instances: A Spot Instance is an instance that uses spare EC2 capacity that is available for a significant discounts compared to the On-Demand price. Learn more at https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/using-spot-instances.html
- AWS Cost and Usage Report: The AWS Cost and Usage Reports (AWS CUR) contains the most comprehensive set of cost and usage data available. You can receive reports that break down your costs by the hour, day, or month, by product or product resource, or by tags that you define yourself. Learn more at https://docs.aws.amazon.com/cur/latest/userguide/what-is-cur.html
- Amazon S3 Bucket: Amazon Simple Storage Service (S3) is an object storage service. A bucket is a container that holds the objects uploaded into S3. It can be helpful to think of a bucket as a folder you use to store files on your computer. Learn more at https://docs.aws.amazon.com/AmazonS3/latest/userguide/Welcome.html
- Amazon CloudWatch: Amazon CloudWatch monitors your Amazon Web Services (AWS) resources and the applications you run on AWS in real time. Learn more at https://docs.aws.amazon.com/AmazonCloudWatch/latest/monitoring/WhatIsCloudWatch.html
- AWS Trusted Advisor: Application that inspects your Amazon Web Services (AWS) environment and provides recommendations to improve performance, security, and cost optimization. Learn more at https://aws.amazon.com/premiumsupport/technology/trusted-advisor/
- Machine-generated checks: Machine-generated checks are checks that are created by a computer or machine, rather than being written out by hand. Learn more at https://aws.amazon.com/blogs/aws/aws-trusted-advisor-new-priority-capability/
- Amazon Elastic Block Store (EBS): Amazon Elastic Block Store (Amazon EBS) provides block level storage volumes for use with EC2 instances. EBS volumes behave like raw, unformatted block devices. Learn more at https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/AmazonEBS.html
- EC2 Instance: An Amazon EC2 instance is a virtual server in Amazon’s Elastic Compute Cloud (EC2) that provides on-demand, scalable computing capacity in the Amazon Web Services (AWS) Cloud. Learn more at https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/concepts.html
- GPU: Graphics Processing Unit (GPU) is a specialized electronic circuit designed to accelerate computer graphics and image processing. Initially developed for graphics manipulation and output, GPUs have evolved to become more generalized computing devices, capable of handling a wide range of applications, including machine learning, scientific image processing, and artificial intelligence training.
- CPU: A Central Processing Unit (CPU), also known as a processor or microprocessor, is the primary component of a computer responsible for executing instructions of a computer program, such as arithmetic, logic, controlling, and input/output operations.
Tasha Penwell is an AWS and Lens Studio Educator, conference presenter, and author. She lives in Ohio with her husband and enjoys photography, exploring new technologies and hosting computer science workshops at high schools. She founded Bytes and Bits when she realized that the traditional education model was creating barriers for the underserved and rural population.
Leave a Reply