GPU vs CPU inference tradeoffs in MLOps - Interactive Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to select the device for inference based on availability.

MLOps

device = 'cuda' if torch.cuda.is_available() else [1]

Drag options to blanks, or click blank then click option'

A'cpu'

B'tpu'

C'gpu'

D'fpga'

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'gpu' as fallback device

Using 'tpu' or 'fpga' without hardware support

When GPU is not available, the CPU is used for inference.

2fill in blank

medium

Complete the code to set batch size for CPU inference to avoid overload.

MLOps

batch_size = [1] if device == 'cpu' else 64

Drag options to blanks, or click blank then click option'

A16

B512

C256

D128

Attempts:

3 left

💡 Hint

Common Mistakes

Using large batch sizes like 128 or 256 on CPU causing slowdowns

Smaller batch sizes like 16 help prevent CPU overload during inference.

3fill in blank

hard

Fix the error in the code that measures inference time on CPU.

MLOps

start = time.time()
output = model(input.to([1]))
end = time.time()

Drag options to blanks, or click blank then click option'

A'gpu'

B'cpu'

C'cuda'

D'tpu'

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'cuda' or 'gpu' when running on CPU causing runtime errors

Input tensor must be moved to CPU device for CPU inference.

4fill in blank

hard

Fill both blanks to create a dictionary showing inference speed tradeoffs.

MLOps

inference_speed = {'CPU': [1], 'GPU': [2]  # in milliseconds

Drag options to blanks, or click blank then click option'

A100

B10

C50

D5

Attempts:

3 left

💡 Hint

Common Mistakes

Swapping CPU and GPU speeds

Using equal speeds for both devices

CPU inference is slower (100 ms) compared to GPU (5 ms) for the same model.

5fill in blank

hard

Fill all three blanks to filter models suitable for CPU inference with low memory.

MLOps

suitable_models = {m: mem for m, mem in models.items() if mem [1] 4 and 'light' [2] m and mem [3] 1}

Drag options to blanks, or click blank then click option'

A<=

Bin

C>=

D==

Attempts:

3 left

💡 Hint

Common Mistakes

Using wrong comparison operators

Using '==' instead of 'in' for substring check

We select models with memory less or equal 4GB, name containing 'light', and memory greater or equal 1GB.

Practice

1. Which of the following is a main advantage of using a GPU over a CPU for machine learning inference?

easy

A. Lower power consumption for small tasks

B. Cheaper hardware cost

C. Better performance on single-threaded tasks

D. Faster processing for large batches of data

Solution

1. Step 1: Understand GPU design for parallelism

   GPUs have many cores designed to handle many operations at once, making them faster for large data batches.

2. Step 2: Compare CPU and GPU strengths

   CPUs are better for single-threaded or small tasks, but GPUs excel at parallel processing, speeding up large inference jobs.

3. Final Answer:

   Faster processing for large batches of data -> Option D

4. Quick Check:

   GPU parallelism = Faster large batch inference [OK]

Hint: GPUs excel at many tasks at once, CPUs at few tasks fast [OK]

Common Mistakes:

• Thinking GPUs always use less power
• Assuming CPUs are cheaper for large-scale inference
• Confusing single-threaded speed with parallel speed

2. Which command correctly runs a TensorFlow model inference on CPU only, ignoring GPUs?

easy

A. CUDA_VISIBLE_DEVICES=0 python inference.py

B. CUDA_VISIBLE_DEVICES='' python inference.py

C. CUDA_VISIBLE_DEVICES=-1 python inference.py

D. CUDA_VISIBLE_DEVICES=all python inference.py

Solution

1. Step 1: Understand CUDA_VISIBLE_DEVICES usage

   Setting CUDA_VISIBLE_DEVICES to an empty string disables GPU visibility, forcing CPU usage.

2. Step 2: Check each option's effect

   CUDA_VISIBLE_DEVICES='' python inference.py disables GPUs correctly; others either select GPUs or use invalid values.

3. Final Answer:

   CUDA_VISIBLE_DEVICES='' python inference.py -> Option B

4. Quick Check:

   Empty CUDA_VISIBLE_DEVICES disables GPU [OK]

Hint: Empty CUDA_VISIBLE_DEVICES means no GPU used [OK]

Common Mistakes:

• Using 0 disables only GPU 0, not all GPUs
• Using -1 is invalid for CUDA_VISIBLE_DEVICES
• Assuming 'all' enables all GPUs but not disables

3. Given this Python snippet for inference timing:

import time
start = time.time()
# Run model inference here
end = time.time()
print(round(end - start, 2))

If GPU inference takes 0.05 seconds and CPU inference takes 0.5 seconds, what will be printed when running on CPU?

medium

A. 0.05

B. 50.0

C. 0.5

D. 5.0

Solution

1. Step 1: Understand timing code output

   The code prints the elapsed time rounded to 2 decimals, so it shows seconds taken.

2. Step 2: Match CPU inference time to output

   CPU inference takes 0.5 seconds, so the printed output is 0.5.

3. Final Answer:

   0.5 -> Option C

4. Quick Check:

   CPU time = 0.5 seconds printed [OK]

Hint: Printed time matches actual elapsed seconds rounded [OK]

Common Mistakes:

• Confusing milliseconds with seconds
• Choosing GPU time instead of CPU time
• Misreading rounding precision

4. You run inference on a GPU but notice it is slower than CPU. Which fix is most likely to improve GPU inference speed?

medium

A. Increase batch size to better use GPU parallelism

B. Reduce batch size to avoid GPU overload

C. Disable GPU and force CPU usage

D. Use single-threaded CPU mode

Solution

1. Step 1: Identify GPU performance factors

   GPUs perform best with larger batch sizes to utilize many cores efficiently.

2. Step 2: Evaluate options for improving GPU speed

   Increasing batch size improves GPU throughput; reducing batch size or disabling GPU lowers performance.

3. Final Answer:

   Increase batch size to better use GPU parallelism -> Option A

4. Quick Check:

   GPU speed improves with larger batches [OK]

Hint: Bigger batches = better GPU use [OK]

Common Mistakes:

• Thinking smaller batches speed up GPU
• Disabling GPU to fix GPU slowness
• Using single-thread CPU instead of GPU

5. You have a small model and low input volume but a tight budget. Which inference setup is best to minimize cost while maintaining reasonable speed?

hard

A. Use CPU inference with small batch sizes

B. Use GPU inference with large batch sizes

C. Use GPU inference with small batch sizes

D. Use CPU inference with large batch sizes

Solution

1. Step 1: Analyze model size and input volume impact

   Small models and low input do not benefit much from GPU parallelism, so GPU cost is less justified.

2. Step 2: Consider budget and batch size tradeoffs

   CPU inference with small batches reduces cost and matches low volume needs without GPU overhead.

3. Final Answer:

   Use CPU inference with small batch sizes -> Option A

4. Quick Check:

   Small model + low volume + budget = CPU small batch [OK]

Hint: Small model + low volume = CPU for cost savings [OK]

Common Mistakes:

• Choosing GPU despite low volume and budget
• Using large batches on CPU causing delays
• Ignoring cost when selecting GPU