Task Orchestration

Not only can you run a single task in Hatchet, but you can also orchestrate multiple tasks together based on a shape that you define. For example, you can run a task that depends on the output of another task, or you can run a task that waits for a certain condition to be met before running.

1. Declarative Workflow Design (DAGs) — which is a way to declaratively define the sequence and dependencies of tasks in a workflow when you know the dependencies ahead of time.
2. Procedural Child Spawning — which is a way to orchestrate tasks in a workflow when you don’t know the dependencies ahead of time or when the dependencies are dynamic.

Flow Controls

In addition to coordinating the execution of tasks, Hatchet also provides a set of flow control primitives that allow you to orchestrate tasks in a workflow. This allows you to run only what your service can handle at any given time.

1. Worker Slots — which is a way to control the number of tasks that can be executed concurrently on a given compute process.
2. Concurrency Control — which is a global way to control the concurrent execution of tasks based on a specific key.
3. Rate Limiting — which is a global way to control the rate of task execution based on time period.