Why scaling requires different strategies in MLOps - See It in Action

Why Scaling Requires Different Strategies

📖 Scenario: You are working in a team that manages machine learning models in production. As the number of users grows, you notice that the current setup struggles to handle the load. You need to understand why scaling your system requires different strategies depending on the situation.

🎯 Goal: Build a simple Python program that models different scaling strategies and shows how they affect system capacity. This will help you understand why one size does not fit all when scaling machine learning systems.

📋 What You'll Learn

Create a dictionary called systems with three keys: 'single_node', 'horizontal_scaling', and 'vertical_scaling' with values 1, 5, and 2 respectively.

Create a variable called load and set it to 7.

Use a for loop with variables strategy and capacity to iterate over systems.items() and create a new dictionary can_handle_load where each key is the strategy and the value is true if capacity is greater than or equal to load, otherwise false.

Print the can_handle_load dictionary to show which strategies can handle the load.

💡 Why This Matters

🌍 Real World

In real machine learning operations, systems must scale to handle more users or data. Different scaling strategies like adding more machines (horizontal) or upgrading a machine (vertical) have different limits and costs.

💼 Career

Understanding scaling strategies is key for MLOps engineers to keep machine learning models running smoothly as demand grows.

1

Create the initial system capacities

Create a dictionary called systems with these exact entries: 'single_node': 1, 'horizontal_scaling': 5, and 'vertical_scaling': 2.

MLOps

# Create the systems dictionary with exact capacities
# Your code here

Need a hint?

Use curly braces to create a dictionary with the exact keys and values.

2

Set the current load value

Create a variable called load and set it to 7.

MLOps

systems = {'single_node': 1, 'horizontal_scaling': 5, 'vertical_scaling': 2}
# Set the load variable to 7
# Your code here

Need a hint?

Just assign the number 7 to the variable named load.

3

Determine which strategies can handle the load

Use a for loop with variables strategy and capacity to iterate over systems.items(). Inside the loop, create a dictionary called can_handle_load where each key is the strategy and the value is true if capacity is greater than or equal to load, otherwise false.

MLOps

systems = {'single_node': 1, 'horizontal_scaling': 5, 'vertical_scaling': 2}
load = 7
# Create can_handle_load dictionary using a for loop
# Your code here

Need a hint?

Start with an empty dictionary can_handle_load = {}. Then use a for loop to fill it.

4

Print the results

Write print(can_handle_load) to display which scaling strategies can handle the load.

MLOps

systems = {'single_node': 1, 'horizontal_scaling': 5, 'vertical_scaling': 2}
load = 7
can_handle_load = {}
for strategy, capacity in systems.items():
    can_handle_load[strategy] = capacity >= load
# Print the can_handle_load dictionary
# Your code here

Need a hint?

Use the print function to show the dictionary.