Microsoft Workloads on AWS

Increasing sustainability for your Microsoft workloads on AWS

At re:Invent 2021, Amazon Web Services (AWS) launched the sustainability pillar as part of the AWS Well-Architected framework. It focuses on minimizing the environmental impact of running cloud workloads. Key topics include a shared responsibility model for sustainability, understanding impact, and maximizing utilization to minimize required resources and reduce downstream impact.

In this blog post, we will take a deep-dive into the sustainability pillar and discuss how you can apply it to your Microsoft workloads on AWS. We also will look at many of the AWS services that can support and advance your own sustainability efforts.

The relationship between Microsoft workloads and sustainability

For legacy Microsoft workloads such as Windows Server and Microsoft SQL Server (in public cloud or on premises), many of the migration, modernization, and optimization conversations focus on cost reduction. For instance, when you use a smaller Amazon Elastic Compute Cloud (Amazon EC2) instance type that is sized appropriately to match the workload, you not only reduce your compute spend, but you may reduce your licensing costs at the same time. Often, licenses for products like Microsoft SQL Server are based on the number of virtual CPU cores used. If you can use a virtual machine with fewer cores, you can reduce the number of licenses required.

Fewer cores + Right-sized compute/storage = Lower consumption billing and license elimination

AWS normally uses this formula to help customers reduce their overall spend, but the benefits of this exercise extend beyond just cost optimization strategies. For Microsoft workloads on AWS, the sustainability impact has begun to align with corporate sustainability goals and commitments. In the sustainability equation, Microsoft workload optimization and modernization often provides the largest impact outside of full data center migration. This is because optimizing Microsoft workloads focuses on core reduction, compute performance optimization, and options for application modernization.

Where is the best place to start?

Job #1 is to gather and understand the data. Specifically, what is running on-premises and in the cloud, and how is that inventory consuming resources? It may sound like a daunting task, but AWS funds this activity for you through the AWS Optimization and Licensing Assessment (OLA). An OLA first catalogs your machines (physical and virtual), identifying the hardware that is used, the applications that are running, and how the compute nodes perform over time. The output gives a view of over-provisioned servers, core count reduction opportunities, containerization options, and application dependency mapping for migration and modernization. The OLA provides data that can help you reduce core counts for specific applications, such as SQL Server. Reducing the number of cores can significantly lower license and compute spend, as well as reduce overall power consumption and resource utilization.

Describes the phases of the Optimization and Licensing Assessment: Scope Workloads, Collect Data, Analyze, and Plan.

Figure 1: Optimization and Licensing Assessment workflow

Environmental sustainability is a shared responsibility

Describes the shared responsibility model, separating responsibilities between the customer and AWS.

Figure 2: The shared responsibility model – Sustainability Pillar

Sustainability of the cloud

AWS is responsible for the sustainability of the cloud – delivering efficient, shared infrastructure, water stewardship, and sourcing renewable power. Cloud providers have a lower carbon footprint and are more energy efficient than typical on-premises alternatives. They invest in efficient power and cooling technology, operate energy efficient server populations, and achieve high server utilization rates. Studies by 451 Research have shown that the AWS infrastructure is 3.6 times more energy efficient than the median of U.S. enterprise data centers surveyed and up to five times more energy efficient than the average in Europe.

Cloud workloads reduce impact by taking advantage of shared resources, such as networking, power, cooling, and physical facilities. You can migrate your cloud workloads to more efficient technologies as they become available, and use cloud-based services to transform your workloads for better sustainability.

Sustainability in the cloud

Customers are responsible for sustainability in the cloud – optimizing workloads and resource utilization, and minimizing the total resources required to be deployed for workloads. Sustainability in the cloud is a continuous effort, focused primarily on energy usage and efficiency of all components of a workload, by striving for maximum utilization of the provisioned resources, while minimizing the total needed. Efforts include the initial selection of an efficient programming language, adoption of modern algorithms, usage of efficient data storage techniques, deploying correctly sized and efficient compute infrastructure, and minimizing the need for high-powered end-user hardware.

Applying sustainability design principles to your Windows workloads

The sustainability pillar of the well architected framework provides six design principles that can help maximize sustainability and minimize impact when architecting cloud workloads. We will review a few of these design principles and highlight some of the AWS tools and services that can help you achieve sustainability goals.

Maximize utilization

Over-provisioning resources can lead to unnecessary infrastructure cost, low resource utilization, and unneeded energy consumption. You can right-size your Windows Server workloads and implement an efficient design to ensure maximum utilization and energy efficiency of the underlying hardware. For example, you can use AWS Compute Optimizer to help right size Amazon EC2 instance types, Amazon Elastic Block Store (EBS) volumes, and AWS Lambda functions. Compute Optimizer provides recommendations for AWS services by applying the knowledge drawn from Amazon’s own experience of running diverse workloads in the cloud. First, Compute Optimizer analyzes a workload’s configuration and resource utilization to identify dozens of defining characteristics. Then, Compute Optimizer determines how the workload would have performed on various hardware platforms (e.g., Amazon EC2 instance types) or by using different configurations (e.g. Amazon EBS volume IOPS settings, and AWS Lambda function memory sizes) to offer recommendations.

Anticipate and adopt newer, more efficient hardware and software offerings

At AWS, we are constantly innovating on behalf of our customers. You can follow along with all the AWS feature and region expansion announcements on the “What’s New with AWS?” page. This makes it easy to gain awareness of and evaluate newer, more efficient hardware and software offerings when they are released. AWS periodically launches new Amazon EC2 instances with more efficient x86 CPUs. Newer generations offer improved performance, more efficient energy usage, and better price performance. Running Microsoft workloads on more modern Amazon EC2 instances helps decrease AWS data center energy consumption, increase sustainability both in and of the cloud.

Refactoring from .NET Framework to .NET is also beneficial because it enables customers to use Amazon EC2 instances powered by AWS Graviton processorsMigrating from x86 processors to the Arm-based AWS Graviton processors reduces costs and energy consumption. Graviton-based EC2 instances provide the same performance as x86 based EC2 instances while consuming up to 60% less energy.

Use managed services

AWS offers managed cloud services where we assume additional operational responsibility, enabling you to focus your efforts on the activities that bring real value to your customers. Sharing services across a broad customer base helps maximize resource utilization, which reduces the overall amount of infrastructure needed to support cloud workloads.

You can choose from a variety of AWS managed services that support Microsoft workloads, such as AWS Directory Service for Microsoft Active Directory, Amazon Relational Database Service  for SQL Server, Amazon FSx for Windows File Server, and AWS Fargate for Windows containers. You can move to managed services such as AWS Systems Manager from Center Configuration Manager (SCCM) for operations, and to Amazon CloudWatch from System Center Operations Manager (SCOM) for a cloud-based monitoring solution.

This allows you to increase sustainability, be more productive, reduce cost, and eliminate the undifferentiated heavy lifting of infrastructure management.

Reduce the downstream impact of your cloud workloads

A focus on sustainability requires an understanding of the devices and equipment used to consume services, their expected lifecycle, and the financial and sustainability impact of replacing those components. A key element of this is to use modern software patterns. You can combine Distributed Load Testing on AWS with Amazon CloudWatch Application Insights to get a better understanding of how an application performs under high-traffic conditions and use that information to determine how to optimize your application code.

AWS also offers services that help you understand how applications will run on end-user devices. For example, AWS Device Farm can give insights into how an application will perform on a variety of mobile devices.

Conclusion

An increasing number of organizations are setting sustainability targets in response to changes in government regulation, competitive advantage, and customer, employee, and investor demand. Architects, CTOs, developers, and operations team members are seeking ways they can directly contribute to their organization’s sustainability goals. By using these design principles, you can strengthen sustainability outcomes for your Microsoft workloads running on AWS services while still balancing security, cost, performance, reliability, and operational excellence.

Every action you take to reduce resource usage and increase efficiency across your workloads contributes to a reduction in environmental impact and contributes to your organization’s broader sustainability goals.

To begin increasing the sustainability of your Microsoft workloads, the first step is to gather and understand the data. We encourage you to reach out to your AWS account team to learn more about the Optimization and Licensing Assessment (OLA), or you can fill out this form and someone from AWS will contact you. For more information about sustainability in the cloud, visit the AWS sustainability page.

Rodney Underkoffler

Rodney Underkoffler

Rodney is a Senior Solutions Architect at Amazon Web Services, focused on guiding enterprise customers on their cloud journey. He has a background in infrastructure, security, and IT business practices. He is passionate about technology and enjoys building and exploring new solutions and methodologies.

Chase Lindeman

Chase Lindeman

Chase Lindeman is a Senior Microsoft Specialist Solutions Architect at Amazon Web Services who has over 15 years of experience working with Microsoft technologies and specializes in infrastructure subjects such as networking, storage, Active Directory, and SQL Server. He has in-depth AWS knowledge and expertise in running Microsoft workloads on AWS.

John Stasick

John Stasick

John is a Migration & Modernization Sales Leader at Amazon Web Services, helping enterprises build innovative applications and services that differentiate their business and transform their customers' experiences.

Marcio Morales

Marcio Morales

Marcio Morales is a Principal Specialist Solution Architect at Amazon Web Services, helping customers to migrate and modernize their infrastructure into AWS. He is the author of the book "Running Windows Containers on AWS" and a global SME for Windows containers. He helps AWS customers design, build, secure, and optimize Windows container workloads on AWS.