Google Cloud Professional Data Engineer — Question 112

You are working on a niche product in the image recognition domain. Your team has developed a model that is dominated by custom C++ TensorFlow ops your team has implemented. These ops are used inside your main training loop and are performing bulky matrix multiplications. It currently takes up to several days to train a model. You want to decrease this time significantly and keep the cost low by using an accelerator on Google Cloud. What should you do?

Answer options

• A. Use Cloud TPUs without any additional adjustment to your code.
• B. Use Cloud TPUs after implementing GPU kernel support for your customs ops.
• C. Use Cloud GPUs after implementing GPU kernel support for your customs ops.
• D. Stay on CPUs, and increase the size of the cluster you're training your model on.

Correct answer: C

Explanation

The correct answer is C because using Cloud GPUs provides the necessary acceleration for the custom operations while allowing for the implementation of GPU kernel support, which is essential for efficiency. Options A and B are not suitable as they do not address the need for custom ops, and option D is inefficient since it continues to rely on CPUs, which are slower for this type of workload.