Building an enterprise integration architecture using Azure - Brown Bag 1

When you are dealing with an enterprise architecture, integrating different systems in real time/near real time/batch mode is always one of the elephant in the room. Purpose of the article is to somewhat simplify the process and take a dive in the latest trend setters to make an API-integrator's life easy.

For our easy understanding let's divide integration layer into the below components based on:

1. Trigger Points - Synchronous(Real time), Fire and forget (Near real time or Async) and Scheduled.
2. Flow of data - Inbound and Outbound
3. Integration end points - Rest/WCF end points, File based or SQL based
4. Location of 3rd party - On Premise/On Cloud/ Hybrid
5. Throughput and Performance expectations

Available Azure technologies that natively cater to all of the above.

1. Azure Logic apps
2. Azure Functions
3. Azure Service Bus (For near real time implementation)
4. Azure Table/Blob storage (For storing temporary & data operations)
5. Integration accounts (For maintaining transformation logic)
6. Web API (For offloading any extensive business logic)
7. Azure Api Management (For load balancing)

Or you can go all ballistic and create custom components for the same, which will lead to bigger dev & test cycles and more maintainability.

Watch out this space for more detailed analysis on each of these components including cost analysis. Also, feel free to comment if you would want any specific details to be added.