ML model training in Snowflake - Time & Space Complexity

Time Complexity: ML model training in Snowflake

O(n)

Understanding Time Complexity

When training a machine learning model in Snowflake, it is important to understand how the time to complete training changes as the data size grows.

We want to know how the number of operations or work done increases when we use more data for training.

Scenario Under Consideration

Analyze the time complexity of the following operation sequence.

CREATE OR REPLACE MODEL my_model
  OPTIONS(
    model_type = 'linear_regression',
    input_label_cols = ('target')
  ) AS
SELECT * FROM training_data;
    

This sequence creates and trains a linear regression model using all rows from the training_data table.

Identify Repeating Operations

Identify the API calls, resource provisioning, data transfers that repeat.

• Primary operation: Reading each row of training_data to compute model parameters.
• How many times: Once per row in the training_data table.

How Execution Grows With Input

As the number of rows in training_data increases, the work to process each row and update model calculations grows proportionally.

Input Size (n)	Approx. Api Calls/Operations
10	About 10 row reads and calculations
100	About 100 row reads and calculations
1000	About 1000 row reads and calculations

Pattern observation: The work grows directly with the number of rows; doubling rows roughly doubles the work.

Final Time Complexity

Time Complexity: O(n)

This means the time to train the model grows in a straight line with the number of data rows.

Common Mistake

[X] Wrong: "Training time stays the same no matter how much data we use."

[OK] Correct: More data means more rows to read and process, so training takes longer as data grows.

Interview Connect

Understanding how training time grows with data size helps you explain performance and scalability clearly in real-world cloud projects.

Self-Check

"What if we added feature selection before training? How would the time complexity change?"