Skip to content
Draft: this topic has not been reviewed yet. Check facts against the linked AWS documentation before relying on them.

The AWS Well-Architected Framework

Cloud Practitioner · Domain 1 · Cloud ConceptsTask 1.2: Identify design principles of the AWS CloudLevel 118 minDraft
Difficulty
Confidence

How do you know whether a cloud setup is good? The AWS Well-Architected Framework answers that with six pillars, each a different way a system can succeed or fail, plus general design principles that apply to all of them.

The Cloud Practitioner exam mostly asks you to recognize a pillar from a short description, so this topic focuses on what each pillar cares about.

Your workloadOperationalexcellenceSecurityReliabilityPerformanceefficiencyCostoptimizationSustainabilityGeneral design principles
The six pillars hold up a well-architected workload; the general design principles are the foundation under all of them.

The six pillars

Operational excellence

Can we run this system well, see what it is doing, and keep improving it?

  • Do operations as code, for example with scripts and infrastructure as code instead of manual steps.
  • Make small, frequent, reversible changes.
  • Learn from failures and update procedures.

Security

Are data, systems and assets protected?

  • Build a strong identity foundation and grant least privilege.
  • Keep traceability: log and review who did what.
  • Apply security at every layer and encrypt data in transit and at rest.

Reliability

Does the system recover from failure and keep up with demand?

  • Recover automatically from failure.
  • Scale horizontally (many smaller resources instead of one large one).
  • Test recovery procedures, and stop guessing capacity.

Performance efficiency

Does it use the right resources efficiently as needs change?

  • Choose the right resource types and sizes for the job.
  • Use serverless and managed services to avoid managing servers.
  • Go global in minutes and experiment more often.

Cost optimization

Does it deliver business value at the lowest price?

  • Pay only for what you use (a consumption model).
  • Measure efficiency and remove waste, such as idle resources.
  • Stop spending money on undifferentiated heavy lifting.

Sustainability

Does it keep environmental impact as low as possible?

  • Maximize utilization so less hardware is needed.
  • Use managed services, which share infrastructure efficiently.
  • Reduce the amount of data and computing your workload needs.

Sustainability is the newest pillar, so older study material may list only five.

General design principles

These sit underneath all six pillars and explain why the cloud changes how you design:

  • Stop guessing capacity needs. Scale with demand instead of buying for the peak.
  • Test systems at production scale. Create a full-size test environment, then delete it.
  • Automate with experimentation in mind. Infrastructure as code makes it cheap to try and compare designs.
  • Allow for evolutionary architectures. Design so parts can be replaced as needs change.
  • Drive architectures using data. Measure and let the data guide decisions.
  • Improve through game days. Practice failures and big events before they happen for real.

Using the framework

  • The AWS Well-Architected Tool in the console walks you through questions for each pillar, records the risks it finds for a workload, and suggests improvements. It is free to use.
  • Lenses add extra questions for specific kinds of workload, such as serverless applications or machine learning.

Putting it together: reviewing a web app

A team reviews its web app with the Well-Architected Tool and finds:

Finding Pillar
Everyone shares one admin login Security
The database runs in a single Availability Zone Reliability
Large servers sit at 10% CPU all month Cost optimization (and sustainability)
Deployments are done by hand at midnight Operational excellence
The app uses old instance types that are slower and pricier Performance efficiency

Common mix-ups

  • Reliability vs performance efficiency. Reliability is about recovering and staying up. Performance efficiency is about using the right resources efficiently.
  • Cost optimization vs sustainability. Both like removing idle resources. Cost is about money; sustainability is about environmental impact.
  • Operational excellence vs reliability. Operational excellence is about how you run and improve the system; reliability is about the system staying available.

Check yourself

A design spreads a database across multiple Availability Zones so it can fail over automatically. Which pillar does this support most?

Reliability. It helps the system recover from failure and stay available.

A team turns on detailed logging of every API call and reviews it weekly. Which pillar is that?

Security, specifically traceability.

Which pillar was added most recently, and what does it focus on?

Sustainability. It focuses on reducing the environmental impact of running workloads.

Key takeaways

  • Six pillars: operational excellence, security, reliability, performance efficiency, cost optimization, sustainability.
  • General design principles, such as stop guessing capacity and test at production scale, apply to all pillars.
  • The Well-Architected Tool reviews a workload against the pillars for free.

Cloud Study Lab is an independent study site. It is not affiliated with or endorsed by Amazon Web Services. AWS and the names of AWS certifications are trademarks of Amazon.com, Inc. or its affiliates.