Building a DAG of models in dbt - Practice Problems & Coding Challenges

Challenge - 5 Problems

🧠 Conceptual

intermediate

Understanding DAG dependencies in dbt

In dbt, models are organized in a Directed Acyclic Graph (DAG). What does it mean when a model depends on another model in this DAG?

A The dependent model runs before the model it depends on.

B Both models run at the same time regardless of dependency.

C The dependent model runs after the model it depends on.

D The dependent model ignores the model it depends on.

💡 Hint

Think about the order of execution when one model uses data from another.

In a DAG, a model that depends on another must run after the model it depends on to ensure data is available.

❓ data_output

intermediate

Resulting DAG nodes count

Given these dbt models with dependencies:

• model_a (no dependencies)
• model_b depends on model_a
• model_c depends on model_a
• model_d depends on model_b and model_c

How many nodes will the DAG contain after compilation?

A 5

B 3

C 2

D 4

💡 Hint

Count each unique model as one node.

Each model is a node. Here, model_a, model_b, model_c, and model_d make 4 nodes.

❓ Predict Output

advanced

Output of dbt run with model dependencies

Consider this simplified dbt project with models and their dependencies:

models:
  model_x.sql: no dependencies
  model_y.sql: depends on model_x
  model_z.sql: depends on model_y

If you run dbt run --models model_z, which models will be run and in what order?

A model_x, model_y, then model_z run in order.

B Only model_z runs.

C model_z runs first, then model_y and model_x.

D model_y and model_z run, model_x is skipped.

💡 Hint

dbt runs all dependencies of the specified model.

dbt runs dependencies first to ensure data is ready. So model_x runs, then model_y, then model_z.

🔧 Debug

advanced

Identifying a cyclic dependency error

You have these dbt models:

• model_1 depends on model_2
• model_2 depends on model_3
• model_3 depends on model_1

What error will dbt raise when compiling this DAG?

A dbt will raise a missing model error.

B dbt will raise a cyclic dependency error.

C dbt will skip model_3 and run others.

D dbt will run models in any order without error.

💡 Hint

Think about what happens when dependencies loop back to the start.

A DAG cannot have cycles. dbt detects this and raises a cyclic dependency error.

🚀 Application

expert

Optimizing DAG execution with selective model runs

You have a large dbt project with many models. You want to run only models affected by changes in model_sales and its downstream models. Which dbt command achieves this?

A <code>dbt run --models +model_sales</code>

B <code>dbt run --models model_sales+</code>

C <code>dbt run --models model_sales</code>

D <code>dbt run --models +model_sales+</code>

💡 Hint

Use dbt's selector syntax to include downstream models.

The model_sales+ selector runs model_sales and all models that depend on it downstream.