Creating and then deploying your Azure Function to the cloud in your usual way in Visual Studio automatically creates and deploys that function.json file we talked about earlier that contains the configuration/metadata information for your Azure Function that it requires based on the code you just wrote. The cloud runtime environment reads the file when it spins up your Azure Function and then runs the logic in the function itself. All in response to the activated trigger.

Triggers are one kind of binding. But Azure Functions also make use of input and output bindings to allow you to easily connect to your related cloud resources such as data and other resources. Azure Functions 2.0 and later supports a variety of bindings that you can use for inbound or outbound connection to those resources.

Note quickly that all bindings except HTTP and Timer must be registered. .NET functions can access and register bindings just by using NuGet packages. Other functions can use extension bundles to access bindings through configuration settings.

Notice in the function.json file that input and output bindings can use directions both in and out, depending on the type of the binding. And some bindings even support a special direction InOut that is available using the Advanced editor in the Azure portal, but we won't be talking about InOut bindings in this opening course on Azure Functions.

So what kinds of resource bindings are supported in Azure Functions? Well, for input you've got Blob Storage, Cosmos DB, IoT Hub, Microsoft Graph Excel tables, Microsoft Graph OneDrive files, Microsoft Graph events, Microsoft Graph Auth tokens, mobile apps, SignalR, Table storage and even Twilio. And for output bindings, it's all of the above except for the Graph Auth tokens, IoT Hub and SignalR. Plus, Event Grid, Event Hubs, HTTP and webhooks, Microsoft Graph Outlook email, Notification Hubs, Queue Storage, SendGrid, and Service Bus. 

In other words, your ability to create interactions in your Azure Functions between the various parts of your cloud architecture is nearly limitless through the use of input and output bindings.

So let's continue our code example by adding an Output binding to our Azure Function. Suppose that we have a scenario in which whenever our current Timer triggered Azure Function goes off, we want to send a message to our Queue. Let's just hop back into Visual Studio to implement that code.