Azure Durable Functions
In the previous chapter, we discussed how HTTP APIs in Azure Functions provide an alternative hosting model, as well as solve many traditional problems associated with manual hosting.
Do you need to massively parallelize hundreds of millions of tasks on an arbitrary number of physical processes or nodes (fan-out) and then wait for them all to complete (fan-in)? Do you then need the app to wait for a human or other process to review and take action to continue the next step? What if that human is unavailable and you need to timeout the review process instead? This situation of events is easily expressible in code, and we’ll be tackling a subset of this scenario in this chapter.
In this chapter, we will look at the following:
- Expanding on standard APIs to create resilient, stateful workflows
- Differences in debugging these workflows compared to previous chapters
- Interacting and injecting data into live systems using standard HTTP APIs...
