Cost optimization at scale in MLOps - Mini Project: Build & Apply

Cost optimization at scale

📖 Scenario: You work as an MLOps engineer managing cloud resources for machine learning projects. Your goal is to optimize costs by tracking resource usage and identifying expensive resources to reduce waste.

🎯 Goal: Build a simple Python program that stores resource costs, sets a cost threshold, filters resources exceeding the threshold, and prints them out. This helps identify which resources to optimize or shut down.

📋 What You'll Learn

Create a dictionary with resource names and their monthly costs

Add a variable for the cost threshold

Filter resources with costs above the threshold using a dictionary comprehension

Print the filtered expensive resources

💡 Why This Matters

🌍 Real World

Cloud and MLOps engineers often need to monitor and optimize resource costs to save money and improve efficiency.

💼 Career

This project teaches basic cost tracking and filtering skills useful for managing cloud budgets and optimizing machine learning infrastructure.

1

Create resource cost dictionary

Create a dictionary called resource_costs with these exact entries: 'GPU_Cluster': 1200, 'Data_Storage': 300, 'API_Server': 450, 'Model_Training': 900, 'Monitoring': 150

MLOps

# Create the resource_costs dictionary with exact entries
# Your code here

Need a hint?

Use curly braces to create a dictionary with keys as resource names and values as costs.

2

Set cost threshold

Create a variable called cost_threshold and set it to 500

MLOps

resource_costs = {
    'GPU_Cluster': 1200,
    'Data_Storage': 300,
    'API_Server': 450,
    'Model_Training': 900,
    'Monitoring': 150
}
# Create cost_threshold variable and set it to 500
# Your code here

Need a hint?

Just assign the number 500 to the variable named cost_threshold.

3

Filter expensive resources

Create a dictionary called expensive_resources using a dictionary comprehension that includes only entries from resource_costs where the cost is greater than cost_threshold

MLOps

resource_costs = {
    'GPU_Cluster': 1200,
    'Data_Storage': 300,
    'API_Server': 450,
    'Model_Training': 900,
    'Monitoring': 150
}
cost_threshold = 500
# Create expensive_resources dictionary with costs above cost_threshold
# Your code here

Need a hint?

Use a dictionary comprehension with for name, cost in resource_costs.items() and an if condition.

4

Print expensive resources

Write a print statement to display the expensive_resources dictionary

MLOps

resource_costs = {
    'GPU_Cluster': 1200,
    'Data_Storage': 300,
    'API_Server': 450,
    'Model_Training': 900,
    'Monitoring': 150
}
cost_threshold = 500
expensive_resources = {name: cost for name, cost in resource_costs.items() if cost > cost_threshold}
# Print the expensive_resources dictionary
# Your code here

Need a hint?

Use print(expensive_resources) to show the filtered dictionary.