AWS Operational Excellence is the Well-Architected pillar for running and improving cloud workloads. It connects automation, operational metrics and failure reviews to business goals.
This guide covers the pillar's design principles and best practice areas. For the broader context, start with the six AWS Well-Architected pillars.
AWS Well-Architected Operational Excellence Pillar Overview
The pillar covers how teams organize, prepare workloads for production, operate them and improve over time. Operational excellence requires both technical practices and clear responsibility for the work.
Understanding AWS Well-Architected Operational Excellence Pillar
Operational Excellence Pillar General Design Principles
AWS's current design guidance extends the earlier focus on operations as code and learning from failures. It includes these principles:
Align teams with business outcomes
Set operational goals that support business priorities. Give teams clear ownership and measures of success.
Use observability to guide action
Use workload telemetry to understand health and performance, then act when business outcomes are at risk.
Automate safely
Treat infrastructure and operational procedures as code. Use guardrails, error thresholds and approvals to control automated actions.
Make Frequent, Small, Reversible Changes
Keep deployments small enough to diagnose and reverse when they fail.
Refine Operations Procedures Frequently
Review procedures as workloads change. Test them regularly and share updates with the people who use them.
Anticipate Failure
Rehearse failure scenarios to test procedures and team responses before a live incident.
Learn from operational events and metrics
Use events, failures and operational data to identify improvements. Share lessons across teams.
Use managed services
Reduce the work your team must operate directly. Document how to support the managed services you use.
Operational Excellence Pillar Best Practice Areas
AWS groups the pillar into four best practice areas:
- Organization: Set priorities, ownership and responsibilities around business objectives.
- Prepare: Design for operational insight and assess readiness before production.
- Operate: Monitor workload health, respond to events and support day-to-day operations.
- Evolve: Use feedback and operational measures to prioritize improvements.
Start by checking whether your team knows who owns each workload, how to detect problems and how to reverse a failed change. Use the results to choose the next operational improvement.
Need Help Today?
The Autimo team can help you navigate the complex cloud security landscape and help build automation and baked-in security into your engineering platforms, leaving your business to focus on delivering value to your customers through your products and solutions.AWS Secure Environment Accelerator (ASEA): historical context
This article originally referenced ASEA. AWS has ended developer support and archived its repository. AWS now recommends Landing Zone Accelerator for new environments with high compliance requirements. See our AWS Landing Zone Accelerator guide for context.
“Without Autimo’s help there is no way we’d have been able to hit the level of productivity we have seen. To be able to forget about the infrastructure and just get on with writing code has been transformative for our team.”
Markus Westerholz CTO, Hero Innovation Group
