NCA GPU Architecture & Fundamentals 4

Question 1

What is the primary advantage of High Bandwidth Memory (HBM) over GDDR memory in data center GPUs?

Accepted Answer

HBM uses a much wider memory bus (1024+ bits), delivering significantly higher bandwidth

Answer

HBM uses a very wide memory bus (1024 bits or more per stack) combined with stacked DRAM dies, achieving memory bandwidths far exceeding GDDR despite lower clock speeds.

Question 2

What is a 'thread block cluster' introduced in NVIDIA Hopper architecture?

Accepted Answer

A collection of thread blocks that can cooperate and share data via distributed shared memory

Answer

Thread block clusters in Hopper allow thread blocks on different SMs within a GPC to cooperate using Distributed Shared Memory, enabling new levels of inter-SM data sharing.

Question 3

What does the NVIDIA GPU's register file store, and why is it significant for performance?

Accepted Answer

It stores per-thread local variables with zero-latency access, enabling fast arithmetic

Answer

The register file stores per-thread variables and provides single-cycle access latency, making register usage critical for achieving peak arithmetic throughput without memory stalls.

Question 4

Why does 'register spilling' hurt GPU kernel performance?

Accepted Answer

It causes excess register values to be stored in local (global) memory, adding high-latency accesses

Answer

When a kernel uses more registers than available per thread, excess values spill to local memory (backed by global DRAM), incurring hundreds of cycles of latency per access.

Question 5

What is the role of the CUDA Cooperative Groups programming model with respect to GPU synchronization?

Accepted Answer

It provides flexible, programmer-defined synchronization scopes beyond a single thread block

Answer

Cooperative Groups extends CUDA synchronization beyond __syncthreads(), allowing developer-defined groups at warp, block, multi-block, or grid scope for more expressive parallel algorithms.

(NCA) NVIDIA-Certified Associate Practice Test

(NCA) NVIDIA-Certified Associate Practice Test

NCA GPU Architecture & Fundamentals 4