Demystifying GPU microarchitecture through microbenchmarking

TLDR

Graphics processors promise more than an order‑of‑magnitude speedup over conventional processors for certain non‑graphics tasks, yet little is known about their internal architecture beyond manufacturer documentation. The study develops a microbenchmark suite to measure the CUDA‑visible architectural characteristics of the Nvidia GT200 (GTX280) GPU. It employs this microbenchmark suite to probe the CUDA‑visible architectural characteristics of the Nvidia GT200 (GTX280) GPU. The measurements reveal undisclosed characteristics of processing elements and memory hierarchies, exposing undocumented features that affect program performance and correctness, and offering insights useful for optimization, analysis, and modeling on this architecture.

Abstract

Graphics processors (GPU) offer the promise of more than an order of magnitude speedup over conventional processors for certain non-graphics computations. Because the GPU is often presented as a C-like abstraction (e.g., Nvidia's CUDA), little is known about the characteristics of the GPU's architecture beyond what the manufacturer has documented. This work develops a microbechmark suite and measures the CUDA-visible architectural characteristics of the Nvidia GT200 (GTX280) GPU. Various undisclosed characteristics of the processing elements and the memory hierarchies are measured. This analysis exposes undocumented features that impact program performance and correctness. These measurements can be useful for improving performance optimization, analysis, and modeling on this architecture and offer additional insight on the decisions made in developing this GPU.