Migrate

Cost optimization doesn't stop after you complete the migration to Cloud. It's essential to seamlessly manage resources and optimize costs on regular basis. Start by closely monitoring your resources throughout both the migration and post-migration stages to ensure smooth transitions. Develop a thoughtful metrics strategy to demystify cloud economics, enabling informed decision-making. Establish budgeting mechanisms to continually monitor cost and usage, and construct user-friendly dashboards with pre-built visualizations for enhanced visibility. Choose the right purchase options and scalable architectures tailored to your specific workloads, and consider a long-term modernization strategy that embraces cost-effective cloud-native services.

MIG-COST-04: What custom monitoring strategies you have put in place to monitor and manage ongoing cost and usage data as you migrate resources to AWS?

Monitoring is an important part of maintaining and managing the cost and usage of AWS resources during and after migration. Customers use different AWS services and tools to successfully manage and optimize their AWS bill.

MIG-COST-BP-4.1: Create a deliberate metrics strategy to help demystify cloud economics

This BP applies to the following best practice areas: Expenditure and usage awareness

Implementation guidance

Suggestion 4.1.1: Regularly evaluate cloud metrics to ensure they meet current business needs.

Start by creating an effective metrics and reporting strategy, then define a set of metrics to measure the implementation of your strategy. See Crafting a robust metrics strategy to quantify your benefits from the cloud for some proven metrics that follow our see, save, plan, and run framework, and can serve as a starting point for any company

MIG-COST-BP-4.2: Monitor spend and limit unintended or unnecessary costs with budgeting and forecasting tools

This BP applies to the following best practice areas: Expenditure and usage awareness

Implementation guidance

Suggestion 4.2.1: Migrating to AWS can likely reduce your total cost of ownership (TCO), but you still need an effective cost control mechanism to make sure you only pay for what you need.

Set up your cost control system, by focusing on the following core principles:

• Budget your spend with custom thresholds.

• Monitor and analyze how your costs progress toward limits.

• Take action to reduce unintended costs.

AWS Budgets gives you the ability to set custom budgets that alert you when your costs or usage (actual or forecasted) exceed your budgeted amount. With the recent launch of AWS Budget actions, you can now preconfigure actions that can trigger the implementation of AWS Identity and Access Management (IAM) policies or service control policies (SCPs). In addition, you can stop target Amazon Elastic Compute Cloud (Amazon EC2) or Amazon RDS instances in your account.

In a multi-account AWS environment, there are two patterns for managing the budget that you can choose from based on your organization's governance structure:

• Centralized budget management, where the budget is set by the management account for all its member accounts

• Decentralized budget management, where the budget is set for individual member accounts by its owners

For more detail, see the following:

• How to manage cost overruns in your AWS multi-account environment – Part 1

• Control developer account costs with AWS CloudFormation and AWS Budgets

• Smart Budgeting Using Lambda and Service Catalog

MIG-COST-BP-4.3: Use AWS Cost Anomaly Detection in Cost Explorer to quickly improve cost controls

This BP applies to the following best practice areas: Expenditure and usage awareness

Implementation guidance

Suggestion 4.3.1: Cost Anomaly Detection is an AWS Cost Management service that uses advanced machine learning to detect anomalous spend and provide contextualized alert notifications through email and Amazon SNS.

Each anomaly includes root cause analysis and direct links to further investigate in Cost Explorer to understand the unexpected usage and its drivers.

The default configuration includes the creation of an AWS services monitor, which tracks charges of most AWS services (see Quotas and restrictions) deployed now, or in the future, by your management and member accounts. It also includes a daily email subscription, which sends an email if any anomaly is detected or ongoing during that specific day. By default, the primary email address associated with the account will receive a daily summary email for any service anomaly detected that is above $100 and exceeds 40% of the expected spend.

MIG-COST-BP-4.4: Use dashboards that provide pre-built visualizations to help you get a detailed view of your AWS usage and costs as you move resources to AWS

This BP applies to the following best practice areas: Expenditure and usage awareness

Implementation guidance

Suggestion 4.4.1: AWS Cost Explorer provides a high-level view of costs and usage, using the same dataset that is used to generate the AWS Cost and Usage Reports. 

To extract resource-level granularity, you can use Amazon Athena queries, which requires familiarity and previous experience to build complex SQL queries.

Having dashboards that provide pre-built visualizations can help you get a detailed view of your AWS usage and costs.

The Cloud Intelligence Dashboards are a collection of Amazon QuickSight dashboards. They offer powerful visuals, in-depth insights, and intuitive querying without having to build complex solutions or share your cost data with third-party companies.

The Cost Intelligence Dashboard, CUDOS Dashboard, Trusted Advisor Organization (TAO) Dashboard, and Trends Dashboard are built on native AWS services. They are inherently secure because the data resides in the organization. They inherit all the features of Amazon QuickSight, including integration with AWS Identity and Access Management, which makes them highly secure, and Amazon QuickSight being a serverless service allows you to pay as you go and scale on demand.

You do not need coding or SQL skills to customize these dashboards. The visualizations include Machine Learning (ML) driven insights, live trends, actionable recommendations, links to relevant blog posts and AWS service documentation that help you make informed business decisions.

MIG-COST-05: How are you ensuring your target infrastructure is optimized for your workloads?

AWS has a lot of options for instance shapes and sizes, purchase options, scaling options, and managed services. Consider how you can leverage these capabilities during and after migration to maximize cost benefits from your cloud infrastructure.

MIG-COST-BP-5.1: Leverage the right purchase options and scalable architecture for your workloads

This BP applies to the following best practice areas: Cost-effective resources

Implementation guidance

Suggestion 5.1.1: Leverage the right purchase model for your workload. 

Selecting the correct purchasing models can greatly reduce the total cost of running a workload. Workloads with fairly consistent resource requirements should consider reserved instances or savings plans to save up to 72 percent on cost compared to the same resources purchased on-demand. On the other hand, some workloads, such as question and answer (QA) environments, may be temporary or intermittent. This type of workload may benefit from Spot Instances, which can be much more cost effective for well-fit workloads compared to comparable on-demand resources. 

In some cases, licensing may have an impact on purchasing models. For instance, some licenses are bound to physical cores and not transferable to an EC2 instance. In this case, it may be more cost effective to purchase a dedicated host to use for the duration of those licenses. For more detail, see EC2 Reservation Models.

Suggestion 5.1.2: Use fine-grained data collected in the assess phase to right-size your infrastructure as you're migrating.

It is unlikely that the resources allocated to your source infrastructure are exactly what your workload is using. The cloud offers more flexibility when provisioning resources and makes it easier to change the size and shape of the provisioned resources. By sizing your target infrastructure to the actual usage, you can see immediate cost savings by migrating to the cloud. 

When selecting your target resources, consider burstable performance instances, which allow you to run a workload at a fraction of the cost of comparable-sized non-burstable instances. These instances are especially cost-effective for non-production or other workloads that are often near-idle. 

Suggestion 5.1.3: Enable autoscaling immediately after migrating when your workload allows it.

Autoscaling can have a significant impact on infrastructure costs, especially for applications that that have large fluctuations in compute capacity needs over time. If you are able to autoscale, consider that when sizing your target infrastructure. Choose instances that are between your minimum and average compute, memory, and disk usage, and make sure your autoscaling rules allow you to scale up to at least the maximum expected usage. 

MIG-COST-BP-5.2: Identify resources during migration that are likely candidates for cost optimizations later

This BP applies to the following best practice areas: Cost-effective resources

Implementation guidance

Suggestion 5.3.1: Tag your resources as you're migrating to make additional cost optimizations later.

It's often not feasible to make every cost optimization during the initial migration. Tag resources as they're being created to distinguish important categories that can help with cost optimization later. For instance, tag production and pre-production instances so you can use the tags later to automate stopping pre-production instances at night.

MIG-COST-06: What cost optimization tools are you leveraging to reduce your cloud spend?

Once you have successfully migrated workloads to AWS, it is critical to choose the right set of AWS cost optimization tools that provide ability to manage, monitor, and track your spend in the cloud. Understanding your cost structure in comprehensive manner and applying it across spectrum of AWS services allows you to implement the right cost optimization solutions in order to optimize your operational cost after migration or modernization.

MIG-COST-BP-6.1: Use automation to re-evaluate your compute usage periodically

This question applies to best practice area:  Optimize over time

Implementation guidance

Suggestion 6.1.1: Employ cost-optimization tools built specifically for AWS infrastructure. 

AWS provides comprehensive cost management tools for cost optimization. Most workloads continue to evolve over time. Use combination of cost optimization tools to continue cost-optimization post-migration such as Compute Optimizer, Trusted Advisor, rightsizing recommendations, Savings Plans (SP) and Reserve Instances (RI) reports, Amazon CloudWatch alarms, and Amazon S3 Lens based on various services that are part of your AWS environments. These tools analyze your AWS usage and make simple, actionable suggestions to reduce costs of running your workloads on AWS, while incorporating the latest features and pricing. 

• AWS Trusted Advisor: Provides recommendations that help you follow AWS best practices. Trusted Advisor evaluates your account by using checks. These checks identify ways to optimize your AWS infrastructure, improve security and performance, reduce costs, and monitor service quotas. You can then follow the recommendations to optimize your services and resources.

• AWS Compute Optimizer: Identifies whether your AWS resources are optimal, and offers recommendations to improve cost and performance. It helps us with rightsizing recommendations by avoiding over provisioning and under provisioning.

• Rightsizing recommendations: This feature in AWS Cost Explorer helps you identify cost-saving opportunities by downsizing or terminating instances in Amazon Elastic Compute Cloud (Amazon EC2). Rightsizing recommendations analyze your Amazon EC2 resources and usage to show opportunities for how you can lower your spending.

• SP and RI reports: If you own Savings Plans or Reserved Instances, this helps to understand how much SP or RI to purchase. Once you purchase, it also helps you understand the coverage, which is a measure of how many instances are covered out of all your EC2 and RDS instances. 

• Amazon CloudWatch: Provides detailed monitoring of infrastructure components at a line item about different services being consumed and allows to set near real-time alarms for Cloud Watch.

• Amazon S3 Lens: Allows customers to get visibility into their Amazon S3 usage. Amazon S3 Storage Lens is a single place to understand Amazon S3 consumption and provides recommendations for objects that haven't been accessed for long time, different storage tiers, and other storage related metrics.

Follow rightsizing recommendations provided by combination of different AWS native Cloud Financial Management tools.

MIG-COST-07: How are you prioritizing your migrated AWS workloads and further driving cost optimization through modernization?

Migrations don't end when a workload is in AWS. It's important to continue to optimize costs post-migration to get the most value from the cloud. There are various modernization pathways that allow you to further optimize your cost profile on AWS while delivering innovative solutions, reducing time-to-value and improving customer experiences.

MIG-COST-BP-7.1: Create a plan early to optimize after the initial migration

This question applies to best practice area:  Optimize over time

Implementation guidance

Suggestion 7.1.1: Depending on your specific AWS architecture and services being deployed, there are a number of techniques to further optimize cost both at the infrastructure layer, including refactoring your application to take advantage of modern, cloud-native services. 

• Use newer, cost-efficient compute options available for your workloads: There are three important ways to optimize compute costs, and AWS has tools to help you with all of them. It starts with choosing the right Amazon EC2 purchase model for your workloads, then selecting the right instance to fine tune price and performance, and finally mapping usage to actual demand.

• Use an optimal combination of AWS Managed Services: Depending on your existing AWS architecture, you can re-platform applications to further save on operating costs in the cloud. If you are running a SQL database on Amazon EC2, modernizing to an Amazon RDS managed service can remove lot of undifferentiated heavy-lifting and lower overall total cost of ownership (TCO). 

• Modernization pathways: The four most popular modernization pathways are as follows:

  • Serverless: Helps organizations to build and run applications without provisioning or managing infrastructure. These services, such as AWS Lambda and AWS Fargate, allow organizations to worry less about operational overhead and have a faster time to market. Features like automatic scaling and pay-for-use billing drive business agility and cost-efficiency.

  • Containers: Containers provide a standard way to package an application's code, configurations, and dependencies into a single object. Examples of container services include Amazon Elastic Container Service (ECS) and AWS Elastic Kubernetes Service (EKS).

  • Managed data: A fully managed, purpose-built database service, supporting diverse data models and applications.

  • Managed analytics: A range of services supporting analytics use cases like data lake initiatives, big data processing, real-time analytics, and operational analytics.