Recently I conducted a webinar for Microsoft APAC on “Low-Code/ No-Code API Development with Azure Serverless” it is a technical deep dive. You can still watch this webinar here. Within almost an hour I managed to demonstrate the creation and development of a B2B REST API using Azure Serverless components without writing much code. In a series of blog posts, I am planning to cover what I have done in plain text.
In my webinar I used
- Azure SQL Serverless
- Azure Logic Apps
- Azure API Management
- Azure Function
To ensure the “serverlessness” of the B2B API. So the beauty in this architecture is that the B2B API vendor only need to pay for consumption while ensuring the dynamic scalability of the solution. This B2B API is authenticated with a subscription key, which has to be shared between the vendor and the consumer for monetization purposes and analytics.
I used Azure Function to consume the B2B API authenticated with an Azure AD application, again to ensure the serverlessness.
Why Serverless for B2B API
In most of the cases enterprises or the B2B API vendors need to ensure their API must be reachable at any time but again in we’re seen some edge cases, where the APIs are only consumed infrequently and they hit high amount of traffic in a certain day/week of the month or certain hour of the day. In this case, the API must be available for everyone with dynamic scalability. Also they organizations hosting the B2B API doesn’t really want to pay for the cloud resources when the resources are sitting idle. This are the two reasons to use serverless architecture for B2B APIs, not a coincidence- these two are the main reason to go for serverless architecture in any cases.
What happens here is that the Azure SQL Serverless is used as the persistent storage for the API and Azure Logic App serves the purpose of holding the business logic. The Azure Logic Apps template we’re using in this architecture is a HTTP Request – Response, this uses HTTP trigger to be invoked. This HTTP trigger will be invoked by the Azure API management which accept the public traffic.
In this series, I shall discuss about all these technologies one by one and in next part I shall cover “Azure SQL Serverless”.
In this series, I would take a “Tasks API” as an example to demo the implementation.