A final project for NYCU class "Parallel Programming" Aims to parallelize the "lazy approach" in many theoratical work in dynamic matching.

Read this to know more about the details

pthread.cpp and mpi_impl.cpp are implementation in different frameworks

generate.cpp and generate_sparse.cpp are of test purpose, generates the graph in the format

n m
va_1 vb_1
va_2 vb_2
.
.
.
va_m vb_m
q
d x y

vertices for two sides are independent, both 0 indexed. each query is a modification to the edge set. Starting with a d, d = 0 stands for deletion while d = 1 stands for addition.