//MPIInitialize, the program starts in parallel

MPI_Init(&argc,&argv); //MPI is initialized and the program starts in parallel

MPI_Comm_size(MPI_COMM_WORLD,&group_size);//Get the current process ID

MPI_Comm_rank(MPI_COMM_WORLD,&my_rank);//Gets the total number of running processes

//Each process enters data for the 1 / group_size block of source data

for(int i = 0; i < 576 /group_size; i++)


//Inverse function of the product, 5 minutes of data read and projection, etc., to get local results



if(my_rank!=0)//Non-master process


MPI_Send(tempResult); //The non-master process sends the local projection calculation processing result to the main process }

else //Main process


MPI_Recv( tempResult);//The main process receives the local processing results from each child process

WriteFile();//The main process generates precipitation products

VpwdDelete();//Release memory


MPI_Finalize();// MPI parallelization ends