let's talk about server lists.
First of all, the term server list is a bit confusing
when you hear that you're probably thinking that there are no servers.
This is not the case.
Your code is still executed on a server somewhere in the cloud.
There's a pool of servers that the cloud vendor maintains, and those are used to run your code.
However, the server that executes your code is determined by the cloud platform.
Also, there's no guarantee that the next time your code executes it will be on the same server as before
your code executes as a result of a trigger. This can be a scheduled trigger http trigger or an event trigger.
This gives you the flexibility to implement different solutions. Using server list computing,
for example, you can use the http triggers to implement restful API s that are accessed from mobile devices. Or use the scheduled trigger to run background jobs every hour or the event trigger to raise alerts when a sensor detects higher temperature.
Now because the cloud vendor manages the infrastructure and you have no control over it, there are certain things you need to consider while implementing your functionality.
Normally, serverless applications are considered stateless.
Although each one of the servers that executes your code has a local store attached, you should not use the storage
you want to preserve between executions. For that,
you may write and read data from the local storage while your code is executing. But once the execution is complete, your code will not have any more access to that data.
Also, keep in mind that the execution environment is shared, and storing sensitive data locally
may not be a good idea.
If you need to pass data between executions,
you should use state fel, serve Earless service or use a cloud storage service to store data from your stateless apps.
Another thing that you may need to consider is that the execution time for your code is limited.
as your functions have execution time limits of five minutes by default,
you can increase it to 10 minutes. But this is still not enough for long running processes.
If you need to have a job that runs longer, you should consider using any other of the computer technologies as your offers.
Azure has two services that implement server list computing
the first one is azure functions
as your functions uses a code first approach or imperative deployment approach.
This means that you write custom code that gets executed in the cloud.
You can write your functions in C sharp F sharp
as your functions can be stateless and state fel
state fel azure functions are also called durable functions, and they pass context between the executions, allowing you to track prior activity.
There are about a dozen built in bindings and triggers for azure functions. For example, scheduled in http triggers or event triggers based on storage blob creations event great events or storage two messages.
You can also develop custom bindings.
Azure functions can be developed locally using the azure functions, core tools and extensions For visual studio in V S code,
you can manage azure functions from the portal visual studio V S code or using Power Shell or the azure cli.
The other server list implementation in azure is the logic apps.
Logic apps use a designer first or decorative approach.
You build flows using pre built connectors and processing blocks.
There are more than 200 connectors and processing blocks including enterprise and social services like office 3 65 Dynamics 3 65 Salesforce s, A P, G era, Zendesk, Twitter, Facebook and more.
You can implement custom connectors to
you implement the workflows in azure portal or in visual studio, and can persist those in adjacent format with predefined schema.
This allows you to automate the deployment of workflows.
The deployment itself and management can be done in the portal visual studio Power shell or azure CLI.
The last thing you need to remember about logic apps is that they are state fel and can persist information between executions.
Now you know how you can use server lists in azure.