liveBook · Manning

Endpoint

As explained in Microsoft Azure in Action

An API endpoint is a network-accessible interface at which an application or service receives requests and returns responses.

An API endpoint is a web address for calling a particular function over a network, usually with HTTP. Send it the required arguments, and you get a result. An API groups these callable functions so applications can communicate without a user interface. For example, a partner application might use an endpoint to access an NFT image from a web shop. section 8.1 section 8.6

In Azure, the address is only part of the job: something must handle the request, and callers need to know what to send.

  • Import a Function App into API Management to expose its Azure Functions as endpoints. Select a method such as GET to inspect an endpoint’s URL and required parameters. section 8.4.4
  • Test an endpoint in API Management by entering its template parameter and sending a request. In the NFT-image example, the filename identifies the image, and the Function copies it to a blob container. section 8.4.4
  • To expose a Logic App through API Management, give it an HTTP endpoint. A workflow triggered only by a new blob does not meet that requirement. section 8.4.5
C:/> az rest --method post 
--url "https://banningBooksCompVision.cognitiveservices.azure.com/
vision/v4.0/analyze?visualFeatures=Tags,Description" 
--headers "Ocp-Apim-Subscription-Key=<your-key>" 
"Content-Type=application/json" –body
'{"url":"https://myblob.blob.core.windows.net/covers/cover.jpg"}'

This REST request calls a Computer Vision analysis endpoint directly. The URL selects the analysis operation and requested features; a resource key authenticates the call, and the JSON body supplies an image URL. section 7.3.1

An endpoint can also serve a deployed AI model. After fine-tuning on labeled examples, deploy the model to make it available for real-time inference. Choosing what to deploy means weighing response quality, latency, and cost; monitoring a production model is a further step. section 7.4.4 section 7.4.3

Before calling an endpoint finished, check its dependencies and access. An image-copy request can fail if its linked blob containers do not exist. An HTTP-triggered Logic App also raises the question of authenticating incoming requests—publicly reachable is not the same as ready for everyone to use. section 8.4.4 section 8.4.5

How Microsoft Azure in Action covers it

How deep it goes
Defines an endpoint as a network-accessible address and walks through integration and testing, but leaves deeper API Management configuration out of scope.
Examples
Uses an NFT-image copy request, a Computer Vision REST call, and a deployed recommendation model.
What it assumes
Starts with a plain-language API and endpoint definition, then expects the reader to navigate Azure services and supply request parameters.
What you can do after
Operate/configure it in API Management by importing a Function App, inspecting endpoint parameters, and sending a test request.
How it connects
Connects endpoints to HTTP, APIs, API Management, Azure Functions, Logic Apps, authentication, and Blob Storage.

Also explained in

Compare how 26 books explain endpoint →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Microsoft Azure in Action. The section links point to the parts of the book it draws on.