Data parallelism vs model parallelism in MLOps - Practice Questions

Challenge - 5 Problems

🧠 Conceptual

intermediate

Understanding Data Parallelism

Which statement best describes data parallelism in machine learning training?

A Copying the entire model to multiple devices and training on different data batches simultaneously.

B Splitting the model into parts and running each part on different devices.

C Training the model on a single device with one batch of data at a time.

D Using multiple models to train on the same data sequentially.

💡 Hint

Think about whether the model or the data is split across devices.

Data parallelism means copying the full model to multiple devices and each device trains on a different subset of data at the same time.

🧠 Conceptual

intermediate

Understanding Model Parallelism

Which option correctly explains model parallelism in machine learning?

A Running the same model multiple times sequentially on the same data.

B Training multiple copies of the full model on different data batches.

C Using a single device to train the entire model on all data.

D Splitting the model into smaller parts and running each part on different devices simultaneously.

💡 Hint

Consider whether the model or the data is divided across devices.

Model parallelism splits the model itself into parts, each running on different devices to handle large models that don't fit on one device.

💻 Command Output

advanced

Output of Data Parallelism Setup Command

What is the output of this command when setting up data parallelism with PyTorch's DataParallel on 2 GPUs?

MLOps

import torch
import torch.nn as nn
model = nn.Linear(10, 2)
model = nn.DataParallel(model, device_ids=[0,1])
print(model.device_ids)

A [1, 0]

B [0, 1]

C AttributeError: 'DataParallel' object has no attribute 'device_ids'

D None

💡 Hint

Check the attribute that stores device IDs in DataParallel.

The device_ids attribute lists the GPUs used for data parallelism in the order specified.

❓ Troubleshoot

advanced

Troubleshooting Model Parallelism Memory Error

You split a large model across two GPUs using model parallelism, but get a CUDA out-of-memory error on the first GPU. What is the most likely cause?

A The first GPU is assigned too many layers causing memory overflow.

B The batch size is too small to utilize both GPUs.

C Data parallelism was used instead of model parallelism.

D The GPUs are not connected properly.

💡 Hint

Think about how model parts are distributed and GPU memory limits.

If one GPU holds too many model layers, it can run out of memory even if the other GPU is free.

✅ Best Practice

expert

Choosing Between Data and Model Parallelism

Which scenario best justifies using model parallelism over data parallelism?

A When you want to reduce communication overhead by using a single GPU.

B When you want to speed up training by running multiple copies of the model on different data batches.

C When the model is too large to fit into the memory of a single GPU.

D When the dataset is small and fits into one device memory easily.

💡 Hint

Consider the main limitation that model parallelism solves.

Model parallelism is used when the model size exceeds one device's memory, requiring splitting the model across devices.

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