Integration with PagerDuty and Slack in Apache Airflow - Time & Space Complexity

Time Complexity: Integration with PagerDuty and Slack

O(n)

Understanding Time Complexity

When integrating Airflow with PagerDuty and Slack, it's important to understand how the time to send alerts grows as the number of tasks or alerts increases.

We want to know how the system handles more alerts and notifications as workload grows.

Scenario Under Consideration

Analyze the time complexity of the following Airflow task that sends alerts to PagerDuty and Slack for multiple failed tasks.

from airflow import DAG
from airflow.operators.python import PythonOperator
from datetime import datetime

def send_alerts(failed_tasks):
    for task in failed_tasks:
        send_to_pagerduty(task)
        send_to_slack(task)

def send_to_pagerduty(task):
    pass  # sends alert to PagerDuty

def send_to_slack(task):
    pass  # sends alert to Slack

with DAG('alert_dag', start_date=datetime(2024,1,1)) as dag:
    alert = PythonOperator(
        task_id='send_alerts',
        python_callable=send_alerts,
        op_kwargs={'failed_tasks': ['task1', 'task2', 'task3']}
    )

This code loops over failed tasks and sends alerts to PagerDuty and Slack for each one.

Identify Repeating Operations

Look at what repeats in the code:

• Primary operation: Looping over each failed task to send alerts.
• How many times: Once for each failed task in the list.

How Execution Grows With Input

As the number of failed tasks grows, the number of alert sends grows the same way.

Input Size (n)	Approx. Operations
10	20 (10 PagerDuty + 10 Slack sends)
100	200 (100 PagerDuty + 100 Slack sends)
1000	2000 (1000 PagerDuty + 1000 Slack sends)

Pattern observation: The total alert sends grow directly with the number of failed tasks.

Final Time Complexity

Time Complexity: O(n)

This means the time to send alerts grows linearly with the number of failed tasks.

Common Mistake

[X] Wrong: "Sending alerts to PagerDuty and Slack happens all at once, so time stays the same no matter how many tasks fail."

[OK] Correct: Each alert requires a separate send operation, so more failed tasks mean more sends and more time.

Interview Connect

Understanding how alert sending scales helps you design reliable workflows that handle many failures gracefully.

Self-Check

"What if we batch all failed tasks into a single alert message instead of sending one per task? How would the time complexity change?"