Use of Front Door for web applications in Azure Public Cloud.

Azure Front Door is typically used to ensure that users can access web applications in the event of a regional outage, balance requests between instances and to support rate limiting. It works well with CDN. Azure Front Door focuses on global load balancing and site-acceleration and Azure CDN offers static content caching and acceleration. By bringing security with threat protection and advanced OWASP capabilities, Azure Front Door makes CDN a remarkable cloud-native solution. 

By itself, Azure Front Door enables us to define, manage, and monitor global routing for our web traffic by optimizing the best performance and instant global failover for high availability. With Front Door, we can transform our global (multi-region) consumer and enterprise applications into robust, high-performance, and personalized modern applications, APIs, and content that reaches a global audience with Azure. Azure Front Door works at Layer 7 of HTTP/HTTPS layer and uses anycast protocol with split TCP and Microsoft’s global network for improving global connectivity. 

Front Door can be used to guarantee business continuity and disaster recovery. This provides robustness to web applications and function applications such that the state can be recovered after a user or application error, regional data center outage, or unplanned disruptions. By its nature of being purely logic, application and app services have varying options to target recovery of different scopes and levels, some of which are compared below:  

  

·                 Concepts to Understand  

Primary region and secondary region: Two regions are used to achieve higher availability and the application is deployed to each region. The designated primary receives traffic normally and the secondary receives traffic on failover.  

  

Front Door can be configured for priority routing which sends traffic to primary region until it becomes unavailable and routes traffic to secondary instead. Front Door has both routing configuration and health probes to monitor the health of each backend. Ideally, health probes should check for critical dependencies such as apps, queues, and databases.  

  

Geo-replication: is configured for storage accounts, SQL databases and Cosmos DB.  

  

  

·                 Configuration patterns

  

There are several general approaches to achieve high availability across regions which include:  

·                 Active/Passive with hot standby: Traffic goes to one region, while the other is running and ready to accept connections. The other is usually allocated in a different region and is always running.  

·                 Active/Passive with cold standby: Traffic goes to one region while the other waits on cold standby. The secondary region isn’t allocated until needed for a failover. This is less cost but takes longer.  

·                 Active/active – both regions are active, and load balanced equally. If one of the regions becomes unavailable, it is taken out of rotation.