AWS Best Practices - How to Achieve a Well Architected Framework
I'm often asked, "How should I monitor my Amazon Web Services infrastructure? What AWS products and features should I be using?" These two questions can be difficult to answer, but with the help of AWS's top Solutions Architects, you can ask your team a set of key questions to ensure you've built AWS best practices into your stack.
Amazon Web Service’s Well Architected Framework combines the lessons learned from thousands of architectural reviews with years of experience of some of the best Cloud Architects in the world. The result is a set of General Design Principles, and the Five Pillars:
This introductory blog post focuses on the General Design Principles of the framework. The rest of the series will address how to achieve the Five Pillars with native AWS functionality, and how to augment that functionality with Sumo Logic.
Sumo Logic integrates with AWS to collect and monitor both the logs and metrics generated by your infrastructure, and applies advanced machine learning capabilities that reduces the time to identify and resolve issues in your environment.
General Design Principles of AWS Best Practices
The General Design Principles of the Well Architected Framework are:
Stop guessing your capacity needs
Increase or decrease your capacity at will, and automate with tools like AWS’s Auto Scaling Groups
Send Sumo Logic your logs and CloudWatch Metrics to understand when and why Auto Scaling Events are occurring
Test systems at production scale
Quickly deploy a production-scale environment, then tear it down when you’re done testing
Add a Sumo Logic collector image to your EC2 AMIs, or deploy our Docker Container on your host to capture all events and statistics of your production-scale test
Automate to make architectural experimentation easier
The Sumo Logic GitHub integration shows who is changing your infrastructure-as-code repository, or any other code repo you’d like to monitor
Allow for evolutionary architectures
AWS allows you to decouple your applications so that one component failure won't affect the other components of your product or service. Take advantage of offerings like SQS or use Lambda and go serverless!
While you're at it, take advantage of the Sumo Logic Lambda integration. Visualize operational and performance trends like memory and duration usage by function version or alias.
Use data to guide your architecture, and aggregate this data into CloudWatch Log Groups and S3 Buckets
Send your S3 data or CloudWatch Logs to Sumo Logic, then apply advanced machine learning and statistical analysis for enhanced monitoring and alerting
Improve through game days
Simulate ‘game days’ and use heavy loads to stress-test your architecture
Use LogCompare to expose the differences in how your systems perform across time or environment
The next piece of the Well Architected Framework is the Security Pillar. Explore further AWS best practices in security when building your cloud infrastructure.
Complete visibility for DevSecOps
Reduce downtime and move from reactive to proactive monitoring.