======HPC School and Competition 26 June to 2 July 2016 Port Elizabeth======



**Nelson Mandela Metropolitan University (NMMU)**

=====About=====

The **HPC Winter School** is the CHPC's flagship course in //parallel programming// for high performance computing covering:

  * Introduction to high performance computing hardware, systems and techniques
  * Multi-core programming with OpenMP
  * Using message passing (MPI) on cluster super computers
  * High performance co-processors
  * Rapid prototyping for HPC with python



=====APPLICATION=====

  * Applications for The 2016 HPC Winter School are **closed**.




=====Programme Overview=====

Note: **Times changed.**

^  ^  Sun 26 Jun  ^  Monday 27 Jun  ^  Tuesday 28 Jun  ^  Wednesday 29 Jun  ^  Thursday 30 Jun  ^  Friday 1 Jul  ^  Saturday 2 Jul  ^
^ 08:00  |  |  Registration  |  //breakfast* from 08:00 to 09:00//  ||||  Evaluation 1  |
^ 09:30  |  |  Lecture 1  |  Lecture 3  |  Lecture 5  |  Lecture 7  |  Lecture 9  |  //break//  |
^ 11:00  |  |  //break//  |||||   Evaluation 2\\ 10:00--11:30\\ //Departures//\\ //from 12:00//   |
^ 11:30  |  |  Lecture 2  |  Lecture 4  |  Lecture 6  |  Lecture 8  |  Lecture 10  |:::|
^ 13:00  |  |  //Lunch at NMMU Conference Centre//  |||||   //Lunch//   |
^ 14:00  |  |  Practical 1  |  Practical 3  |  Practical 5  |  Practical 7  |  Practical 9  |  Evaluation 3  |
^ 15:30  |  |  //break//  |||||  //Departures//  |
^ 16:00  |  //Arrivals//  |  Practical 2  |  Practical 4  |  Practical 6  |  Practical 8  |  Practical 10  |  Evaluation 4  |
^ 17:30  |  |  //end of day//  |||||  |
^ 18:00  |  //dinner* from 18:00 to 20:00//  ||||||  |


** *Note: Breakfast and dinner only for delegates staying in //Protea// residence.**

====HPC School in Parallel Programming====

===Lectures 1 & 2===

Foundation: Introduction to HPC architectures and systems.  The Linux programming environment.  Job submission and control.

===Lectures 3 & 4===

OpenMP: Programming shared memory (SMP) systems with OpenMP.

===Lectures 5 & 6===

MPI part I: Parallel programming with the message passing library (MPI) for distributed memory systems.

===Lectures 7 & 8===

MPI part II: Design of parallel programs for clusters.  Advanced MPI features including parallel IO and hybrid programming.

===Lectures 9 & 10===

Accelerators: HPC programming on Nvidia GPUs with CUDA.

===Evaluation===

A short practical test.


====Student Cluster Compeitition====

===Lectures 1 & 2===


===Lectures 3 & 4===


===Lectures 5 & 6===


===Lectures 7 & 8===


===Lectures 9 & 10===

===Evaluation===

Note: each team is evaluated, and some teams will have their evaluation after lunch.