• A project containing three labs for a graduate course CS C267: Applications of Parallel Computers, Spring 2023, UC Berkeley
• Official website and lab requirements: CS267, see HW 2-1, HW 2-2, and HW 2-3
• Email me for codes, which are set to private in Github

Contributions

• Implemented an $O(N)$ time particle simulation program given a naive $O(N^2)$ implementation in C++.
• Sped up the program for a shared memory model with OpenMP and a distributed memory model with MPI.
• Parallelized the program on an A100 GPU using CUDA and achieved a 311x speedup over the serial version.
• Report of three homeworks:

CS267_HW2.1.pdf

CS267group6_hw2_2.pdf

CS267group6_hw2_3.pdf