This class will provide a foundation for:

• Developing parallel-programming skills in OpenMP and MPI
• Understanding performance aspects of the AMD & Intel Architecture
• Using batch system: SGE and LSF
• Optimizing code
• Code debugging

Labs for the training will include access to TACC machines for:

• Compiling and running example codes
• Porting your own codes
• Discussion and hands-on assistance from instructors
• Collaborative discussions with colleagues who have interest in parallel computing
• Tours of HPC facilities where appropriate

Technical Requirements

Please Note: We will be unable to provide laptops for the labs for this training class.

Therefore, students should bring their own laptop in order to participate in labs. If you can bring a laptop, we ask that you consider sharing with another student who could not bring one.

You will need a secure shell client on your laptop to access Ranger.

If you are using Windows, you will need a third party ssh client like putty, http://www.chiark.greenend.org.uk/~sgtatham/putty/download.html.

If you have a Mac or Linux laptop, you can use the ssh client provided on these platforms.

Accommodations

Attendees traveling to the training are responsible for arranging and paying for their own travel, daily expenses, and hotel accommodations.

Please send replies and inquiries to garza@tacc.utexas.edu.

Agenda

Day 1: 8:15 am - 5:00 pm
Time	Day 1
8:15 am	Registration
8:30 am	Welcome & TACC Overview
8:45 am	Introduction to Parallel Computing
9:45 am	Break
10:00 am	Rocks/Linux (x86_64) Programming Env.
11:00 am	Lab
12:00 pm	Lunch
1:00 pm	Intro to Programming with OpenMP
2:00 pm	Lab
3:00 pm	Break
3:15 pm	Programming with MPI
4:15 pm	Lab

Day 2: 8:30 am - 5:00 pm
Time	Day 2
8:30 am	HPC Hardware Overview
9:30 am	Hybrid Programming with OpenMP & MPI
10:30 am	Break
10:45 pm	Lab
12:00 pm	Lunch
1:00 pm	Optimization and Scalability for HPC
2:30 pm	Lab
3:00 pm	Break
3:15 pm	Profiling and Debugging
4:00 pm	Lab