Challenge #13: benchmarking different saxpy proble, sizes - jtristan123/HW-for-AI-ML-ECE-410 GitHub Wiki
benchmarking different saxpy proble, sizes
time to get our black leather jackets!!
Goals:
- set up CUDA dev denvironment
- modify CUDA code
- profile and optimize the implementation
- compare and visualize the performance
WHAT IS CUDA?
soo cuda is to program NVIDIA GPUS to run calculations at the same time in parallel
what is SAXPY problems?
AX + Y its math for the GPU to compute
Tasks:
set up CUDA development environment first time doing this lets see how it goes good thing I have a GPU 1660ti, (dreaming of a 50 series)
after speading hours on setting up the evnment
I GOT CUDA ON VSCODE!!
Testing Results:
Testing it and execution time: From 2^N, N = 15, 16, ... 25
Bar Plot:
Question: Why is the low 2^15 slower than the rest?