Monday September 7, 2015
This problem challenges you to create an MPI parallel version of an existing serial C code.
The starter code for this problem is the solution code to a previous challenge problem, "Diffusion, Limited Aggregation in C and MPI." You may wish to familiarize yourself with that problem first (http://hpcuniversity.org/students/weeklyChallenge/106/). The starter code is available at the bottom of the challenge in the "Diffusion, Limited Aggregation in C and X11 solution zip file."
Your task is to create an MPI version of the program that divides the particles among processes. The processes do the calculations for getting stuck and moving randomly on their subsets of particles in parallel.
In this new version of the program, "free" (blue) particles are allowed to move on top of each other.
Only the MPI process with rank 0 should run the X11 code. The MPI process with rank 0 should not do any calculations for getting stuck or moving randomly.
Your task is to create an MPI version of the program that divides the particles among processes. The processes do the calculations for getting stuck and moving randomly on their subsets of particles in parallel.
In this new version of the program, "free" (blue) particles are allowed to move on top of each other.
Only the MPI process with rank 0 should run the X11 code. The MPI process with rank 0 should not do any calculations for getting stuck or moving randomly.
Show solution
Challenge Resources:
Diffusion Limited Aggregation in C and X11 solution zip file
—
Solution to the Diffusion Limited Aggregation in C and X11 challenge problem.
Diffusion Limited Aggregation in MPI solution zip file
—
Solution to the Diffusion Limited Aggregation in MPI challenge problem.
©1994-2024
|
Shodor
|
Privacy Policy
|
NSDL
|
XSEDE
|
Blue Waters
|
ACM SIGHPC
|
|
|
|
|
|
XSEDE Code of Conduct
|
Not Logged In. Login