Task trees and dependencies in Snowflake - Step-by-Step Execution

Process Flow - Task trees and dependencies

Define Tasks

↓

Set Dependencies

↓

Build Task Tree

↓

Execute Root Task

↓

Check Dependencies

↓

Execute Dependent Task 1

↓

Complete

↓

Execute Dependent Task 2

↓

Complete

↓

Complete Root Task

↓

All Tasks Done

Tasks are defined and linked by dependencies forming a tree. Executing a root task triggers dependent tasks first, ensuring order.

Execution Sample

Snowflake

CREATE TASK task_A
  WAREHOUSE = wh
  AS SELECT 1;

CREATE TASK task_B
  WAREHOUSE = wh
  AFTER task_A
  AS SELECT 2;

Defines two tasks where task_B runs only after task_A completes.

Process Table

Step	Action	Task Executed	Dependency Check	Result
1	Start execution of task_B	task_B	Depends on task_A	Wait for task_A to complete
2	Execute task_A	task_A	No dependencies	task_A runs and completes
3	Re-check task_B dependencies	task_B	task_A completed	task_B runs and completes
4	All tasks done	-	-	Execution finished successfully

💡 All tasks executed respecting dependencies; task_B waited for task_A to finish before running.

Status Tracker

Task	Initial State	After Step 1	After Step 2	After Step 3	Final State
task_A	Not started	Not started	Completed	Completed	Completed
task_B	Not started	Waiting	Waiting	Completed	Completed

Key Moments - 2 Insights

Why does task_B not run immediately when triggered?

Because task_B depends on task_A, it waits until task_A completes first (see execution_table step 1 and 2).

What happens if a dependency task fails?

The dependent task will not run, as dependencies must complete successfully before proceeding (implied by dependency check in execution_table).

Visual Quiz - 3 Questions

Look at the execution_table, what is the state of task_B after step 1?

A Running

B Completed

C Waiting

D Not started

💡 Hint

Check the 'Dependency Check' and 'Result' columns in execution_table row 1.

At which step does task_A complete execution?

A Step 2

B Step 1

C Step 3

D Step 4

💡 Hint

Look at the 'Result' column in execution_table for when task_A runs and completes.

If task_A failed, what would happen to task_B?

A task_B runs anyway

B task_B is skipped

C task_B runs before task_A

D task_B waits indefinitely

💡 Hint

Dependencies must complete successfully before dependent tasks run, see key_moments explanation.

Concept Snapshot

Task trees link tasks with dependencies.
A task runs only after its dependencies complete.
Use AFTER clause to set dependencies.
Executing a root task triggers dependent tasks in order.
Failures in dependencies stop dependent tasks from running.

Full Transcript

In Snowflake, tasks can be linked in a tree structure using dependencies. You define tasks and specify which tasks must finish before others run using the AFTER clause. When you execute a root task, Snowflake checks its dependencies and runs them first if needed. For example, task_B depends on task_A, so task_B waits until task_A finishes. This ensures tasks run in the correct order. If a dependency fails, dependent tasks do not run. This helps manage complex workflows reliably.