Core
GPU Architecture and Execution
A GPU is not a fast CPU. It runs thousands of threads in lockstep groups called warps across many streaming multiprocessors, under the SIMT model, and its real constraint is moving data through a memory hierarchy that spans fast on-chip registers and shared memory down to slow off-chip HBM. Understanding occupancy, coalesced memory access, and warp divergence is what separates a kernel that hits peak throughput from one that leaves 90% of the chip idle.
a free account unlocks the core curriculum tier · no card
Related concepts
Practise this in questions
ML Infrastructure & GPUsWalk me through the GPU memory hierarchy, registers, shared memory, L2, HBM. What lives where and why? →easy · FreeMLOps & ML EngineeringContainerize a GPU ML pipeline with Docker. How do you match CUDA to drivers and keep it reproducible? →medium · PremiumLLM & GenAI FundamentalsExplain FlashAttention: why it never writes the full N x N matrix, and how it speeds up attention without changing the math. →hard · PremiumML Infrastructure & GPUsExplain how the CUDA execution model maps to hardware, grids, blocks, warps, SMs. →easy · FreeML Infrastructure & GPUsWhat is warp divergence and why does it hurt performance? →easy · FreeML Infrastructure & GPUsExplain memory coalescing and shared-memory bank conflicts. How would you fix a kernel that has both? →medium · Free