MPI Hands-On; Sending and Receiving Messages II

1. Sending in a ring. A http://siber.cankaya.edu.tr/ozdogan/ParallelComputing/cfiles/code27.c code27.c that takes data from process zero and sends it to all of the other processes by sending it in a ring.
   • That is, process i should receive the data and send it to process i+1, until the last process is reached.
   • Assume that the data consists of a single integer. Process zero reads the data from the user.

2. The http://siber.cankaya.edu.tr/ozdogan/ParallelComputing/cfiles/code3.c code3.c consists of one receiver process and N-1 sender processes.
   • The sender processes send a message consisting of their process identifier (id) and the total number of processes (ntasks) to the receiver.
   • The receiver process prints out the values it receives in the messeges from the senders.

3. Analyse the example http://siber.cankaya.edu.tr/ozdogan/ParallelComputing/cfiles/code2.c code2 for sending/receiving.

4. Exercise: Write a complete program to make a parallel multiplication of two arrays (a and b) with the size of one million each.
   $$c=\sum_{i=1}^{1000000} a[i]*b[i]$$
   • Fill the arrays with random numbers.
   • Master node should distribute the date to the nodes.
   • Your code should work both for even and odd numbers of compute nodes.
   • Master node should also be involved to the action.
   • There should be element-wise multiplication.
   • Each node also should have a local sum and send to master node their values to added by master node up.
   • Master node should print out the final summation value (c) and also report the local values (c at nodes)with corresponding nodes.