Question 27
UnclassifiedA data center team is evaluating an upgrade from DGX H100 to DGX H200 systems for their large language model training workloads. They need to understand the key hardware differences. What are the primary memory improvements in the H200 GPU compared to the H100?
Correct answer: A
Explanation
The H200 upgrades memory capacity and speed: it has "141GB HBM3e memory with 4.8 TB/s bandwidth," while the H100 has "80GB HBM3 with 3.35 TB/s bandwidth." That means the H200 provides a 76% larger memory pool and about 43% more bandwidth, which helps large language model training handle bigger models and move data faster.
Why each option is right or wrong
A. H200 features 141GB HBM3e memory with 4.8 TB/s bandwidth, compared to H100's 80GB HBM3 with 3.35 TB/s bandwidth - a 76% capacity increase and 43% bandwidth improvement
The H200’s distinguishing hardware change is its memory subsystem: NVIDIA specifies 141GB of HBM3e on H200 versus 80GB of HBM3 on H100, and the memory bandwidth rises from 3.35 TB/s to 4.8 TB/s. Numerically, that is a 61GB increase in capacity, or about 76% more memory, plus roughly 43% higher bandwidth, which is the relevant comparison for large-model training workloads.
B. H200 doubles the GPU compute cores from H100 while maintaining the same 80GB memory capacity
C. H200 uses the same memory as H100 but adds dedicated inference accelerators
D. H200 reduces memory to 64GB but increases bandwidth to 6.0 TB/s for inference optimization