In order to successfully complete this assignment you need to participate both individually and in groups during class. If you attend class in-person then have one of the instructors check your notebook and sign you out before leaving class on Wednesday April 21. If you are attending asynchronously, turn in your assignment using D2L no later than _11:59pm on Wednesday April 21.

In-Class Assignment: Final Review¶

Yet another picture of a server room

Agenda for today's class (70 minutes)¶

</p>

(70 minutes) Midterm Review

Here are some files you can use to help you study:

1. Final Review¶

Today's class will be open for your questions and comments. Ask me anything! Here is a summary of the many topics we covered.

HPCC Command Line Interface, Serial and Pleasantly Parallel Jobs¶

Logging onto the HPC
Setting your path
Writing a submission script
Reading man and info pages and googling help
Edit a file
Debugging problems
Compiler optimization
Benchmarking code
Using the module system
Tick Trick
Piping information
BASH Block code
File redirection
File permissions
Block memory
Finding commands
Multiple solutions to finding answers

Shared Memory¶

Basic syntax and debugging for Shared memory parallization (OpenMP)
Compiling OpenMP code
Converting Serial Code to OpenMP Code
Submitting OpenMP Jobs to the HPC
Understanding the strengths and weaknesses of OpenMP

Accelerators¶

Basic Syntax and debugging for GPU Programming (CUDA)
Compiling CUDA code
Converting Serial Code to CUDA
Submitting CUDA Jobs to the HPC
Understanding the strengths and weaknesses of CUDA
Understanding alternatives to CUDA
Comparing Serial, Shared Memory and GPU solutions

Shared Network¶

Basic Syntax and debugging for distributed Memory Parallization (MPI)
Compiling MPI code
Converting Serial Code to MPI Code
Submitting MPI Jobs to the HPC
Understanding the Strings and weaknesses of MPI
Compare Serial, Shared Memory, Distributed Memory and GPU solutions

Parallel Programming topics¶

Understanding and fixing Deadlock
Understanding and fixing Race Conditions
Understanding how to calculate speedup and Amdahl's law
Strong vs Weak Scaling
Understanding hybrid jobs
Estimating scaling of a job
Checkpoint / Restart.
Being able to figure out a never seen before parallel program or library
Planning a large scale workflow.

Congratulations, we're done!¶

If you attend class in-person then have one of the instructors check your notebook and sign you out before leaving class. If you are attending asynchronously, turn in your assignment using D2L.

Course Resources:¶

Written by Dr. Dirk Colbry, Michigan State University
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.