Link to this document's Jupyter Notebook

In order to successfully complete this assignment you must do the required reading, watch the provided videos and complete all instructions. The embedded survey form must be entirely filled out and submitted on or before 11:59pm on Thursday March 16. Students must come to class the next day prepared to discuss the material covered in this assignment.

Pre-Class Assignment: MPI Collective Communication¶

1. Pi revisited¶

Recall that we are trying to compute the integral

$$ \int_0^1 \frac{4}{1+x^2} dx, $$

by approximation using quadrature. In the code cmse401-s23/wave_project/MPI_1D/pi.cpp this is done in the lines

for (int i = istart; i <=iend ; i++)
        x(i) = x_min + h*(i-1);

    double rsum = 0.0;

    for (int i = istart; i <=iend ; i++)
        rsum = rsum + h*integrand(x(i));

✅ DO THIS: Make sure you understand how istart and iend are computed for each rank.

✅ DO THIS: Write out at least one speciffic example, say size=3 and M=10.

In the code we are adding up the Riemann sums from each processor by a sequence of send and recieves. As mentioned in class this only provides the rank=0 processor with the correct global sum.

✅ DO THIS: Add (essentially the reversed communication pattern) send/recv communications so that rank=0 shares the correct result with all processes.

Read the tutorials on broadcast, Scatter / Gather and Reduce

✅ DO THIS: How can you use scatter to inform every other process of the global value. Implement this.

✅ DO THIS: How can MPI_allreduce be used to replace the send/recv combinations? Implement this.

2. Assignment wrap up¶

Please fill out the form that appears when you run the code below. You must completely fill this out in order to receive credits for the assignment!

Direct Link to Google Form

If you have trouble with the embedded form, please make sure you log on with your MSU google account at googleapps.msu.edu and then click on the direct link above.

✅ QUESTION: Summarize what you did in this assignment.