This project demonstrates the performance benefits of using multithreading to compute the sum of a 2D array with a size of 2^16 x 2^16. The project includes two implementations: block-wise and cyclic thread distribution. Performance measurements are obtained for varying block sizes and the results are presented in the form of GFlops graphs. The project aims to showcase the advantages of parallel computing and performance optimization. Note *: when compiling,use -lpthread command with gcc to ensure there are no errors.