Data parallelism vs model parallelism in MLOps - Performance Comparison

Time Complexity: Data parallelism vs model parallelism

O(n)

Understanding Time Complexity

When training machine learning models, we often split work to speed things up. This can be done by splitting data or splitting the model itself.

We want to understand how the time to train changes as we increase data or model size using these two methods.

Scenario Under Consideration

Analyze the time complexity of these simplified parallel training steps.

for each batch in data_batches:  # data parallelism
    send batch to each worker
    worker computes forward and backward pass
    gather gradients and update model

# model parallelism example
split model into parts
for each input batch:
    pass data through model parts sequentially on different devices
    compute gradients and update parts
    

This code shows two ways to split training: by data batches or by model parts.

Identify Repeating Operations

Look at what repeats and costs time:

• Primary operation: Forward and backward passes over data or model parts.
• How many times: For data parallelism, once per data batch per worker; for model parallelism, once per model part sequentially per batch.

How Execution Grows With Input

As data size grows, data parallelism splits batches across workers, so time per batch stays similar but total work grows linearly.

Input Size (n batches)	Approx. Operations
10	10 forward/backward passes split across workers
100	100 forward/backward passes split across workers
1000	1000 forward/backward passes split across workers

For model parallelism, as model size grows, the number of sequential parts grows, increasing time per batch roughly linearly with model parts.

Final Time Complexity

Time Complexity: O(n)

This means training time grows roughly in direct proportion to the number of data batches or model parts processed.

Common Mistake

[X] Wrong: "Splitting data or model always makes training twice as fast when doubling workers or parts."

[OK] Correct: Communication overhead and sequential steps in model parallelism limit speed gains, so doubling resources does not always halve time.

Interview Connect

Understanding how splitting work affects training time helps you explain trade-offs in real projects. It shows you can think about scaling and efficiency clearly.

Self-Check

What if we combined data and model parallelism? How would the time complexity change?

Practice

1. What is the main difference between data parallelism and model parallelism in machine learning training?

easy

A. Data parallelism splits the data across workers, while model parallelism splits the model across workers.

B. Data parallelism splits the model across workers, while model parallelism splits the data across workers.

C. Data parallelism uses only one worker, model parallelism uses multiple workers.

D. Data parallelism trains different models, model parallelism trains the same model multiple times.

Solution

1. Step 1: Understand data parallelism

   Data parallelism means dividing the input data into parts and sending each part to a different worker. Each worker runs the full model on its data part.

2. Step 2: Understand model parallelism

   Model parallelism means splitting the model itself into parts and assigning each part to a different worker. The data flows through these parts sequentially.

3. Final Answer:

   Data parallelism splits the data across workers, while model parallelism splits the model across workers. -> Option A

4. Quick Check:

   Data vs Model split [OK]

Hint: Data parallelism splits data; model parallelism splits model [OK]

Common Mistakes:

• Confusing which is split: data or model
• Thinking both split data only
• Assuming model parallelism uses one worker

2. Which of the following is the correct way to describe data parallelism in a distributed training setup?

easy

A. The data is duplicated on one worker and processed sequentially.

B. Each worker trains a different part of the model on the full dataset.

C. The model is split into layers, each trained by a different worker on the full data.

D. Each worker trains the full model on a subset of the data.

Solution

1. Step 1: Analyze data parallelism setup

   In data parallelism, the full model is copied to each worker. Each worker trains on a different subset of the data.

2. Step 2: Evaluate options

   Each worker trains the full model on a subset of the data. correctly states that each worker trains the full model on a subset of data. Other options describe model splitting or incorrect data handling.

3. Final Answer:

   Each worker trains the full model on a subset of the data. -> Option D

4. Quick Check:

   Full model + data subset [OK]

Hint: Data parallelism = full model per worker, split data [OK]

Common Mistakes:

• Thinking model is split in data parallelism
• Assuming data is duplicated on one worker
• Confusing model layers with data chunks

3. Consider a model split into 3 parts for model parallelism across 3 workers. If input data batch size is 90, how is the data processed?

medium

A. Each worker processes 30 data samples independently on the full model.

B. All 90 samples flow sequentially through the 3 model parts on different workers.

C. Each worker processes all 90 samples on its model part independently.

D. The data is split into 3 parts, each processed by a different worker on the full model.

Solution

1. Step 1: Understand model parallelism data flow

   In model parallelism, the model is split into parts on different workers. The full data batch flows through these parts sequentially.

2. Step 2: Analyze data processing

   All 90 samples pass through the first model part on worker 1, then output flows to worker 2's model part, and so on.

3. Final Answer:

   All 90 samples flow sequentially through the 3 model parts on different workers. -> Option B

4. Quick Check:

   Model split, data flows through [OK]

Hint: Model parallelism splits model; data flows through all parts [OK]

Common Mistakes:

• Assuming data is split in model parallelism
• Thinking each worker processes full data independently
• Confusing data parallelism with model parallelism

4. You tried to implement model parallelism but noticed workers are idle waiting for data. What is the likely cause?

medium

A. Model parts are not connected properly causing data flow delays.

B. Data is not being split correctly across workers.

C. Each worker is running the full model on the full data.

D. Data parallelism was used instead of model parallelism.

Solution

1. Step 1: Identify symptoms of idle workers in model parallelism

   Idle workers waiting for data usually mean data flow between model parts is blocked or delayed.

2. Step 2: Analyze model part connections

   If model parts are not connected properly, data cannot flow smoothly, causing some workers to wait.

3. Final Answer:

   Model parts are not connected properly causing data flow delays. -> Option A

4. Quick Check:

   Idle workers = broken model part connections [OK]

Hint: Idle workers? Check model part connections in model parallelism [OK]

Common Mistakes:

• Blaming data splitting in model parallelism
• Confusing full model runs with model splitting
• Mixing up data and model parallelism issues

5. You have a very large model that does not fit into one GPU memory. Which approach is best to train it efficiently?

hard

A. Use data parallelism by splitting data across GPUs, each with full model copy.

B. Train the model on CPU only to avoid GPU memory limits.

C. Use model parallelism by splitting the model across GPUs, each handling part of the model.

D. Reduce batch size and train on a single GPU.

Solution

1. Step 1: Understand GPU memory limits

   If the model is too large to fit in one GPU, copying full model to each GPU (data parallelism) is not possible.

2. Step 2: Choose model parallelism

   Splitting the model across GPUs allows each GPU to hold only a part of the model, enabling training of large models.

3. Final Answer:

   Use model parallelism by splitting the model across GPUs, each handling part of the model. -> Option C

4. Quick Check:

   Large model fits by splitting model [OK]

Hint: Large model? Split model across GPUs (model parallelism) [OK]

Common Mistakes:

• Trying data parallelism with too large model
• Ignoring GPU memory limits
• Reducing batch size instead of splitting model