Quiz 34 Question 15 of 20

You are reviewing the hardware inventory for your company's deep learning data center, which includes several NVIDIA DGX A100 systems equipped with the 80GB GPU variant. When sizing a model for distributed training across a single DGX A100 node, what is the correct memory allocation structure and total capacity?

Select an answer to reveal the explanation.

Motivation