Mobilize
The next step in preparing your workforce and resources to migrate your enterprise at scale is to break down the mobilize activities into different workstreams. Although the goal of the mobilize phase is the migration of business applications, most prescriptive guidance and answers on achieving your sustainability goals are found here.
| MIG-SUS-03: How do you define and optimize cloud resources during migration so that you become more energy efficient by minimizing idle resources? |
|---|
As a part of your migration planning, one of the important tasks is to define the key workload performance matrix based on your assessment. This plays the significant part in deciding how you would like to migrate the target workload, instead of replicating as-is on-premises configuration. Optimizing cloud resources by removing idle resources helps lower carbon emissions without compromising business requirements.
MIG-SUS-BP-3.1: Focus on efficiency across all aspects of infrastructure
For example, during migration, verify that you use only the required resources, instead of trying to match with source on-premises capacity. This BP applies to the following best practice areas: Alignment to demand
Implementation guidance
Suggestion 3.1.1: Review the on-premises capacity to plan the workload requirements for your target environment.
During migration assessment, define the workload performance metrics for client requests. To optimize cloud resources, take advantage of elasticity in the cloud so you can meet the increasing demand of the migrating workload. As part of your assessment, review and analyze the following:
-
How to respond to the overall demand, rate of change, and required response time, which could potentially help to minimize the environmental impact. Implement dynamic scaling and automation practice aligning to your SLAs to remove excess capacity and assign only needed capacity.
-
Identify redundancy, underutilization, and potential decommission targets, and plan how you can consolidate the redundant content, scale down underutilized resources, and decommission unused assets.
Suggestion 3.1.2: Use proven workflow templates to migrate enterprise applications.
The way you plan your migration can help you identify the automation opportunity to improve the efficiency during migration process. For example:
-
You can use Migration Hub Orchestrator templates to create a migration workflow that can be customized to fit your unique migration requirements, instead of manually performing all the tasks.
-
You can leverage services like AWS Control Tower
to get started. Control Tower helps you set up a multi-account environment and automate the creation of AWS accounts with built-in governance.
Suggestion 3.1.3: Evaluate your migrated workload to consider and configure auto scaling mechanism.
AWS Auto Scaling monitors your applications and automatically adjusts capacity to maintain steady, predictable performance. Define the metrics that have the most relevance to your application's performance to meet changes in demand, and ensure that workloads can scale down quickly and easily during periods of low user load.
Suggestion 3.1.4: Define and update service-level agreements (SLAs).
-
Review and optimize your workload service-level agreements (SLA) based on your sustainability goals to minimize the resources required to support your workload, while continuing to meet business needs. Consider your SLA requirements as part of your design and architecture as well.
-
Define and update SLAs of the migrating workload, such as:
-
Availability or data retention periods, to minimize the number of resources required to support your workload. For more detail, see SUS02-BP02 Align SLAs with sustainability goals
-
Power off the workload during non-functional period. You can use Instance Scheduler on AWS to do this, which automates the starting and stopping of Amazon Elastic Compute Cloud (Amazon EC2) and Amazon Relational Database Service (Amazon RDS) instances.
-
Identify if use of a messaging component can lead to relaxed service-level requirements that can be met with fewer resources. Employ batching requests, where possible, for optimal use of resources. For more detail, see SUS03-BP01 Optimize software and architecture for asynchronous and scheduled jobs.
-
For more detail, see the following:
| MIG-SUS-04: Do you consider sustainability when selecting and prioritizing applications for migration and modernization? |
|---|
Have sustainability in mind when deciding which applications to migrate and modernize. To do this, first identify metrics that can act as a stand-in for your application's sustainability. Then, use these metrics to initiate migration and modernization projects that improve the application's sustainability.
MIG-SUS-BP-4.1: Adopt metrics that can signal the sustainability of your application
This BP applies to the following best practice areas: Process and culture
Adopt metrics to understand what you have provisioned and how those resources are consumed. Evaluate potential improvements, and estimate their potential impact, the cost to implement, and the associated risks. Measure improvements over time to study trends and the impacts of any migration and modernization initiatives.
Implementation guidance
Suggestion 4.1.1: Adopt sustainability metrics relevant to the application.
Understand the resources provisioned by your application to complete a unit of work. Leverage monitoring tools to define proxy metrics
You can also add other sustainability attributes that signal the efficiency of your application. An example of this is the version of your operating systems, runtimes, middleware, libraries, and applications. Keeping your workloads up-to-date can improve workload efficiency, and capturing this information keeps stakeholder informed. Another example of this is an indicator to note if you are using Graviton-based instances improve the performance efficiency of your application. For more detail, see AWS Well-Architected Framework - Sustainability Pillar.
For more detail, see
Measure and track cloud efficiency with sustainability proxy metrics, Part II: Establish a metrics pipeline
Suggestion 4.1.2: Introduce organizational dashboards to share sustainability metrics with stakeholders.
Leverage automation to report, visualize, and enforce sustainability metrics for your application. Introduce organizational dashboards for sustainability that can be shared with application owners and other stakeholders, including key organizational functions such as a Cloud Center of Excellence (CCoE) and Application Review Board (ARB). Make sustainability metrics a part of your architectural decisions.
MIG-SUS-BP-4.2: Include sustainability metrics in the application portfolio analysis to drive migration and modernization initiatives
This BP applies to the following best practice areas: Process and culture
Include sustainability metrics when scoping and prioritizing applications for migrations. Sustainability may be excluded if alignment from migration goals and organizational goals is missing.
Implementation guidance
Suggestion 4.2.1: Include sustainability metrics in the application portfolio analysis.
Establish a sustainability improvement process and adopt methods that can rapidly introduce sustainability improvements and keep your workloads up-to-date.
Including sustainability metrics in the application portfolio analysis assures that relevant data is captured early and is included in architecture and implementation considerations. These metrics capture the business and technical value of retiring applications, and prioritize rightsizing and application scaling to meet infrequent demand.
| MIG-SUS-05: How do you implement efficient workload design to support your sustainability goals? |
|---|
Compute and storage services make up the foundation of many customers, which brings great potential for workload design consideration that can improve the energy efficiency of the migrating workload.
MIG-SUS-BP-5.1: Implement efficient workload design by leveraging the underlying infrastructure.
For example, right-size the workload for the target state before migrating to minimize idle resources, and to avoid over provisioned capacity. This BP applies to the following best practice areas: Hardware and services
Implementation guidance
Suggestion 5.1.1: Select the most efficient hardware and services for your workload migration.
Amazon EC2 provides a wide selection of instance types
Suggestion 5.1.2: Gain insight into workload performance.
Gain insight into workload performance metrics like CPU utilization, memory utilization, network utilization, and disk and usage patterns to perform the right alignment of the cloud resource.
-
Key matrices to consider are the following:
-
If the workload is idle for a long time, it's a good sign to investigate if this workload can retire instead of migrating it.
-
Understand your average CPU and memory utilization, and identify resources that are underutilized. Rightsize those instances to reduce your carbon footprint.
-
Identify the network and storage performance, such as I/OPS, and size for the target workload. Analyze the overall demand, rate of change, and required response time to rightsize the throttle or buffer required.
-
For fault-tolerant, flexible, and stateless workloads that can trade-off with minimal interruption, adopt Spot Instances in your design.
-
Evaluate your migrated workload continually using AWS Compute Optimizer and AWS Trusted Advisor
, and proactively make rightsizing adjustments on your workload.
-
Suggestion 5.1.3: Consider managed services in your workload design.
Remove the need for you to run and maintain physical servers, as AWS operates at scale and is responsible for their efficient operation. For example, instead of migrating your virtual machine or container to Amazon ECS running on Amazon EC2, consider using a service like Amazon Fargate, where you can run containers without having to manage the servers, or package and deploy Lambda functions as container images.
Containerize and migrate existing applications using , for migrating and modernizing Java and .NET web applications into container format. Use managed services to operate more efficiently in the cloud.
Suggestion 5.1.4: Configure your instance tenancy, which defines how EC2 instances are distributed across physical hardware.
Tenancy provides the opportunity to further optimize the workload. Migrating to shared instances instead of migrating to dedicated underutilized instances or hosts can be more beneficial when working towards the sustainability goal.
For more detail, see the following:
| MIG-SUS-06: How do you take advantage of software and architecture patterns for workloads you are going to migrate to support your sustainability goals? |
|---|
Software and architecture patterns can be used to influence utilization of resources while migrating workloads. The aim is to use patterns that maximize utilization so that resources consumed for the workloads are minimized. Idling of resources due to user behavior or nature of workload can also be minimized by applying appropriate software and architecture patterns. on-premises workloads are not designed to take advantage of AWS cloud features that can optimize utilization of resources, like autoscaling. These workloads can have multiple instances of software or services running at multiple locations or are overprovisioned all the time to cater to peak demand. Some workloads are running all the time, even when not in use. Using the 7 Rs can optimize resource usage. Adopt patterns and architecture to consolidate underutilized components to increase overall utilization. Retire components that are no longer required.
MIG-SUS-BP-6.1: Identify environments and workloads that can be consolidated or retired
This BP applies to the following best practice areas: Software and architecture
Implementation guidance
Suggestion 6.1.1: Consolidate environments and workloads in the AWS Cloud.
When moving workloads from multiple on-premises environments or in merger and acquisition cases, consolidating environments on AWS leads to optimal usage of resources and eliminates duplicate functionality. Identify workloads during the assess stage that can be eliminated or consolidated. Identify multiple instances of services or applications running at multiple locations on-premises.
Retire workloads that are not used. Use automated tools such as Migration Evaluator
MIG-SUS-BP-6.2: Identify workloads that can use efficient software and architecture patterns to maintain consistent high utilization of deployed resources
This BP applies to the following best practice areas: Software and architecture
Implementation guidance
Suggestion 6.2.1: Optimize software and architecture for asynchronous and scheduled jobs.
Identify applications and workloads that can benefit from software and architecture patterns to maintain consistently-high utilization of deployed resources while migrating applications.
While migrating applications, evaluate use of integration patterns to scale the processing independently of the receiving of messages, which reduces resource utilization. Identify if use of a messaging component can lead to relaxed service level requirements that can be met with fewer resources. Employ batching requests where possible for optimal use of resources, as batching provides consistent usage. Scheduling the batch jobs reduces idle time for resources. For detail, see SUS03-BP01 Optimize software and architecture for asynchronous and scheduled jobs.
MIG-SUS-BP-6.3: Analyze your data access patterns and data lifecycle processes, and evaluate how you can become more efficient and sustainable in your data management
This BP applies to the following best practice areas: Software and architecture
Implementation guidance
Storing and accessing data efficiently, in addition to reducing idle storage resources, results in a more efficient and sustainable architecture. When migrating data, understand how data is used within your workload, consumed by your users, transferred, and stored. Use software patterns and architectures that best support data access and storage to minimize the compute, networking, and storage resources required to support the workload.
Suggestion 6.3.1: Define and implement a data lifecycle process for data in your object store.
Design a data lifecycle management process based on your data access patterns, observed in your on-premises facility. That process either removes data that is no longer required or archives data into less resource-intensive storage. While migrating data, implement a data lifecycle policy. For more detail, see Best practice 15.4 – Implement data retention processes to remove redundant data from your analytics environment.
Suggestion 6.3.2: Evaluate use of columnar data formats and compression.
While migrating data, evaluate if columnar data formats like Parquet and ORC can be used. These formats require less storage capacity
-
Parquet consumes up to six times less storage in Amazon S3 compared to text formats. This is because of features such as column-wise compression, different encodings, or compression based on data type
. -
You can improve performance and reduce query costs of Amazon Athena
by 30–90 percent by compressing, partitioning, and converting your data into columnar formats. Using columnar data formats and compressions reduces the amount of data scanned.
MIG-SUS-BP-6.4: Understand and influence business requirements, and optimize areas of code to reach your sustainability goals
This BP applies to the following best practice areas: Software and architecture
Understanding your sustainability goals is the first step to focusing on the factors needed to meet those goals. Defining such criteria involves adopting metrics that can be used to measure and evaluate your current sustainability posture, report progress against goals, and accelerate improvements. By analyzing the current environmental impact of the underlying cloud-based infrastructure, you can quantify the tradeoffs and changes required to meet your sustainability objectives.
Implementation guidance
Suggestion 6.4.1: Define criteria to measure and understand your sustainability impact after your migration.
Post-migration, you can use sustainability proxy metrics for your monitoring scenarios. Proxy metrics
Example proxy metrics include vCPU minutes for compute, GBs provisioned for storage, and GBs transferred for network traffic. Proxy metrics combined with business metrics can define sustainability KPIs, which can be used to drive sustainability optimizations while keeping business needs in focus. One example would be to measure vCPU minutes per transaction and define an improvement goal to minimize this metric. Business stakeholders would have to weigh the cost, as reducing vCPUs could ultimately become detrimental to delivering on business needs. When running workloads in AWS, the change in these measured resources correlates with a similar change in cost (except as noted in the following), making overall infrastructure spend a useful proxy metric.
By agreeing on a set of sustainability metrics, the architect team can evaluate different technical approaches to reduce environmental impact.
For more detail, see the following:
Suggestion
6.4.2: Consider using
Amazon CodeWhisperer
