10.5 Pillar Four - Cost Optimization

Use the Cost Optimization pillar to reduce your costs to a minimum and use those savings for other parts of your business. A cost optimized system allows you to pay the lowest price possible while still achieving your business objectives.

Design Principles

Transparently attribute expenditure
Use managed services to reduce cost of ownership
Trade capital expense for operating expense
Benefit from economies of scale
Stop spending money on data center operations

Definition

Cost Optimization in the cloud consists of 4 areas:

Matched supply and demand
- Best practices: Try to optimally align supply with demand. Don't over provision or under provision, instead as demand grows, so should your supply of compute resources. Think of things like auto-scaling which scale with demand. Similarly in a serverless context, use services such as Lambda that only execute (or respond) when a request (demand) comes in.
- Services such as CloudWatch can also help you keep track as to what your demand is.
- Questions to ask yourself:
  - How do you make sure your capacity matches but doesn't substantially exceed what you need?
  - How are you optimizing your usage of AWS services?
- AWS Key services
  - Auto-scaling
Cost-effective resources
- Best practices: Using the correct instance type can be key to cost savings. For example, you might have a reporting process that is running on a t2-micro and it takes 7 hours to complete. That same process could be run on an m4.2-xlarge in a manner of minutes. The result remains the same but the t2-micro is more expensive because it ran for longer.
- A well architected system will use the most cost efficient resources to reach the end business goal.
- Questions to ask yourself:
  - Have you selected the appropriate resources types to meet your cost targets?
  - Have you selected the appropriate pricing model to meet your cost targets?
  - Are there managed services (higher level services than Amazon EC2, Amazon EBS, and Amazon S3) that you can use to improve your ROI (Return On Investment)?
- AWS Key services:
  - EC2 (Reserved instances), AWS Trusted Advisor.
Expenditure awareness
- Best practices: With cloud you no longer have to go out and get quotes on physical servers, choose a supplier, have those resources delivered, installed, made available etc. You can provision things within seconds, however this comes with its own issues. Many organizations have different teams, each with their own AWS accounts. Being aware of what each team is spending and where is crucial to any well architected system. You can use cost allocation tags to track this, billing alerts as well as consolidated billing.
- Questions to ask yourself:
  - What access controls and procedures do you have in place to govern AWS costs?
  - How are you monitoring usage and spending?
  - How do you decommission resources that you no longer need, or stop resources that are temporarily not needed? (e.g. how do you stop the bill of an employee who has left the company but you've set up a cross account billing/consolidated billing)
  - How do you consider data-transfer charges when designing your architecture?
- AWS Key services:
  - CloudWatch Alarms, SNS.
Optimizing over time
- Best practices: AWS moves FAST. There are hundreds of new services (and potentially 1000 new services this yeaer). A service that you choose yesterday may not be the beset service to be using today. For example, consider MySQL RDS, Aurora was launched at re:invent 2014 and is now out of preview. Aurora may be a better option now for your business because of its performance and redundancy. You should keep track of the changes made to AWS and constantly re-evaluate your existing architecture. You can do this by subscribing to the AWS blog and by using services such as Trusted Advisor.
- Questions to ask yourself:
  - How do you manage and/or consider the adoption of new services? (e.g. you need to subscribe to the AWS blog)
- AWS Key services:
  - AWS blog, AWS Trusted Advisor.

Exam Tips:

Cost Optimization in the cloud consists of 4 areas and their questions.

Previous10.4 Pillar Three - Performance Next10.6 Pillar Five - Operational Excellence

Last updated 5 years ago

Was this helpful?

Definition

Cost Optimization in the cloud consists of 4 areas:

Matched supply and demand

Best practices: Try to optimally align supply with demand. Don't over provision or under provision, instead as demand grows, so should your supply of compute resources. Think of things like auto-scaling which scale with demand. Similarly in a serverless context, use services such as Lambda that only execute (or respond) when a request (demand) comes in.
Services such as CloudWatch can also help you keep track as to what your demand is.
Questions to ask yourself:
- How do you make sure your capacity matches but doesn't substantially exceed what you need?
- How are you optimizing your usage of AWS services?
AWS Key services
- Auto-scaling

Cost-effective resources

Best practices: Using the correct instance type can be key to cost savings. For example, you might have a reporting process that is running on a t2-micro and it takes 7 hours to complete. That same process could be run on an m4.2-xlarge in a manner of minutes. The result remains the same but the t2-micro is more expensive because it ran for longer.
A well architected system will use the most cost efficient resources to reach the end business goal.
Questions to ask yourself:
- Have you selected the appropriate resources types to meet your cost targets?
- Have you selected the appropriate pricing model to meet your cost targets?
- Are there managed services (higher level services than Amazon EC2, Amazon EBS, and Amazon S3) that you can use to improve your ROI (Return On Investment)?
AWS Key services:
- EC2 (Reserved instances), AWS Trusted Advisor.

Expenditure awareness

Best practices: With cloud you no longer have to go out and get quotes on physical servers, choose a supplier, have those resources delivered, installed, made available etc. You can provision things within seconds, however this comes with its own issues. Many organizations have different teams, each with their own AWS accounts. Being aware of what each team is spending and where is crucial to any well architected system. You can use cost allocation tags to track this, billing alerts as well as consolidated billing.
Questions to ask yourself:
- What access controls and procedures do you have in place to govern AWS costs?
- How are you monitoring usage and spending?
- How do you decommission resources that you no longer need, or stop resources that are temporarily not needed? (e.g. how do you stop the bill of an employee who has left the company but you've set up a cross account billing/consolidated billing)
- How do you consider data-transfer charges when designing your architecture?
AWS Key services:
- CloudWatch Alarms, SNS.

Optimizing over time

Best practices: AWS moves FAST. There are hundreds of new services (and potentially 1000 new services this yeaer). A service that you choose yesterday may not be the beset service to be using today. For example, consider MySQL RDS, Aurora was launched at re:invent 2014 and is now out of preview. Aurora may be a better option now for your business because of its performance and redundancy. You should keep track of the changes made to AWS and constantly re-evaluate your existing architecture. You can do this by subscribing to the AWS blog and by using services such as Trusted Advisor.
Questions to ask yourself:
- How do you manage and/or consider the adoption of new services? (e.g. you need to subscribe to the AWS blog)
AWS Key services:
- AWS blog, AWS Trusted Advisor.