Everything fails, all the time
Ricardo Martins
![Ricardo Martins]()
Ricardo Martins
Architecting Cloud Solutions for Scalability, Resiliency, & Innovation | Tech Mentor | Cloud Evangelist
This quote is from Werner Vogels, the VP & CTO at Amazon.com. And this is what you must have in mind when designing for cloud.
Build your workload in cloud doesn't mean that your workload never will fail. Everything fails, all the time.
When you start to draft your architecture, consider that all can fail. This is what will ensure your availability. Keep in mind:
Design for failure and nothing fails.
So I've curated some important links to you be aware on how build your Azure architecture using our best practices:
- Reliability with Microsoft Azure
- SLA for Azure Services
- Microsoft Azure Well-Architected Framework
- The reliability pillar
- Resiliency checklist for Azure Services
- Failure mode analysis for Azure applications
- Recover from a region-wide service disruption
- Data Management for Reliability
- Chaos engineering
- Testing Azure applications for resiliency and availability
- Availability options for VMs in Azure
- Azure regions decision guide
- Building solutions for High Availability using Availability Zones
- Build great solutions with the Microsoft Azure Well-Architected Framework (Microsoft Learn Learning Path)
Sample Architectures:
- Run a web application in multiple Azure regions for high availability
- Run an N-tier application in multiple Azure regions for high availability
- Multitier web application built for high availability and disaster recovery on Azure
- Highly scalable and secure WordPress website
Principal Solutions Architect | Empowering clients to build informed and responsible decision cultures
4 年Nice, architecture. For the persistent data, SQL and Cosmos, are they both geo replicated or are you just managing that with Cosmos?