AWS Well-Architected Framework: Performance Efficiency

Tal Shladovsky
June 22, 2023
min. read
Related Resource(s)
No items found.


The Performance Efficiency pillar of the AWS Well-Architected Framework focuses on optimizing cloud compute to meet performance requirements.

This pillar consists of design principles and best practices on cloud computing performance, including: 

  • Use of advanced computing technologies including AI, ML and serverless,
  • Experimentation,
  • Global architectures,
  • Aligning the design of software to the cloud technologies it is running on.


The Performance Efficiency pillar of the AWS Well-Architected Framework is focused on optimizing the use of cloud computing resources to meet performance requirements and efficiently deliver business value. This includes understanding workload requirements, selecting appropriate resource types, and monitoring performance to identify opportunities for optimization and cost savings.

Design Principles

The Performance Efficiency pillar consists of the following five design principles:

  • Democratize advanced technologies: This principle focuses on leveraging advanced technologies, such as machine learning and artificial intelligence, to optimize performance and reduce costs. It encourages teams to learn about and experiment with new technologies and tools to drive innovation and increase efficiency.
  • Go global in minutes: This principle emphasizes the importance of designing architectures that can operate globally, with minimal effort and complexity. This includes using cloud-native services, such as Content Delivery Networks (CDNs) and edge computing, to distribute content and workloads closer to end-users, and reducing latency and improving performance.
  • Use serverless architectures: This principle focuses on the use of serverless architectures to eliminate the need for servers and infrastructure management, and reduce costs while increasing scalability and performance. This includes using serverless computing services, such as AWS Lambda, to run code and perform functions without needing to provision or manage servers.
  • Experiment more often: This principle encourages teams to experiment and innovate more frequently, to identify new ways to optimize performance and reduce costs. This includes using techniques such as A/B testing and blue/green deployments, to test and validate changes and gather data to inform future decisions.
  • Consider mechanical sympathy: This principle involves understanding the underlying hardware and infrastructure on which the workload is running and designing the workload to take advantage of the underlying hardware to optimize performance. For example, optimizing the workload to take advantage of the available memory and storage capacity, and utilizing the underlying infrastructure features such as caching and load balancing to maximize efficiency.

Best Practices

The Performance Efficiency pillar consists of the following four best practices:

  • Selection: This area of the Performance Efficiency pillar focuses on selecting the most appropriate compute and storage resources for the workload based on factors such as performance requirements, usage patterns, and cost. It includes evaluating various resource types, such as instances, containers, and serverless computing, and selecting the optimal resource type based on workload characteristics.
  • Review: This area involves regularly reviewing the performance of the workload and the use of computing resources to identify opportunities for optimization and cost savings. This includes conducting periodic reviews of resource utilization, identifying bottlenecks and inefficiencies, and making changes to the architecture or resource utilization to improve performance.
  • Monitoring: This area focuses on implementing effective monitoring and alerting to detect and diagnose performance issues and identify opportunities for optimization. It includes using monitoring tools and techniques to measure key performance metrics, such as latency, throughput, and error rates, and setting up automated alerts and notifications to trigger actions based on performance thresholds.
  • Trade-offs: This area involves making trade-offs between performance, cost, and other factors, such as security and compliance, when designing and optimizing the workload. It includes considering factors such as availability requirements, scalability, and performance, and making informed decisions about resource utilization and architectural design based on these factors. This may involve making trade-offs between performance and cost or between performance and other requirements to achieve the optimal balance.


The Performance Efficiency pillar plays a vital role in optimizing the performance and cost-effectiveness of applications running on the AWS platform. By following the best practices and principles outlined in this pillar, organizations can ensure that their systems operate at peak efficiency, deliver exceptional user experiences, and effectively utilize available resources.
Adopting a data-driven approach is crucial for achieving and maintaining performance efficiency. By analyzing access patterns and making informed trade-offs, organizations can optimize their systems for higher performance. Conducting thorough reviews based on benchmarks and load tests enables the selection of appropriate resource types and configurations, resulting in optimal performance and cost optimization.   
Treating infrastructure as code allows for rapid and safe evolution of the architecture. By leveraging tools like AWS CloudFormation and infrastructure-as-code principles, organizations can automate the deployment and management of resources, enabling agility and scalability while maintaining consistency and reducing the risk of errors. The ability to make fact-based decisions about the architecture is key to performance efficiency. By leveraging data and monitoring tools like AWS CloudWatch and AWS X-Ray, organizations can gain insights into system behavior, identify performance bottlenecks, and make informed decisions to optimize performance and resource allocation.
Combining active and passive monitoring ensures that the performance of the architecture remains consistent over time. Proactive monitoring, alerting, and automated scaling using services such as AWS Auto Scaling and Amazon CloudWatch Alarms enable organizations to dynamically adjust resources based on demand, maintaining optimal performance while controlling costs.

By prioritizing performance efficiency within the AWS Well-Architected Framework, organizations can maximize the value they derive from their AWS resources, enhance the user experience, and optimize costs. Through careful analysis, continuous monitoring, and the adoption of best practices, organizations can build high-performing, scalable, and cost-efficient architectures that meet the evolving demands of their applications and users.

What's new