Do you really know how GPUs work? (Ep. 259)
Data Science at Home
00:00
Exploring the Intricacies of CPU and GPU Architectures and Execution Models
This chapter explores the design disparities between CPUs and GPUs, focusing on sequential and parallel instruction execution. It discusses architectural variations, concepts like instruction latency, caches, ALUs, and cores, as well as GPU execution features such as threads, blocks, warps, occupancy, and resource allocation, stressing the importance of hardware knowledge for writing efficient code.
Transcript
Play full episode
