Gain insights into using HPC systems and supercomputers to solve large computational problems. Explore HPC components, software stacks, job schedulers, and parallel programming, including Open MP, MPI, and GPU coding.

Learn to use High Performance Computing (HPC) Systems  and solve large computational problems.  This course assumes basic familiarity with the Bash command line environment found on GNU/Linux and other Unix-like environments. This course is of the first of its kind, should be your second step from my previous Educative course "Learn to Analyze Text Data in Bash Shell and Linux"

You'll learn:
- Intro to HPC Systems and Supercomputers
- HPC system's basic components
- HPC software stack
- HPC job schedulers and batch systems (PBS and Slurm with demos)
- Introduction to parallel programming (Open MP, MPI and GPU coding)

Video demos are included, I think it is a must have course for students, researchers, big data analysts, and developers. It will help you to utilize High Performance Computing systems at your institution. Non-interactive course. More details at:  www.scientificprogramming.io/learn-hpc

Ajay Shenoy

This course demystifies HPC and introduces you to the essential tools required to schedule and run parallel programs. Easy to understand.

India

Learn to Use HPC Systems and Supercomputers

A **work-sharing construct** divides the execution of the enclosed code region among the members of the team that encounter it, but do not launch new threads. There is no implied barrier upon entry to a work-sharing construct, however there is an implied barrier at the end of a work sharing construct. 

There are three type of constructs:

### Do/for-loop

Shares iterations of a loop across the team. Represents a type of "data parallelism".

````
#pragma omp for [clause ...]  newline 
                schedule (type [,chunk]) 
                ordered
                private (list) 
                firstprivate (list) 
                lastprivate (list) 
                shared (list) 
                reduction (operator: list) 
                collapse (n) 
                nowait 

   for_loop

````

### Sections

Breaks work into separate, discrete sections. Each section is executed by a thread. Can be used to implement a type of "functional parallelism".

````
#pragma omp sections [clause ...]  newline 
                     private (list) 
                     firstprivate (list) 
                     lastprivate (list) 
                     reduction (operator: list) 
                     nowait
  {

  #pragma omp section   newline 

     structured_block

  #pragma omp section   newline 

     structured_block

  }
````

### Single

Serializes a section of code.

````
#pragma omp single [clause ...]  newline 
                   private (list) 
                   firstprivate (list) 
                   nowait

     structured_block
````


A **work-sharing construct** divides the execution of the enclosed code region among the members of the team that encounter it, but do not launch new threads. There is no implied barrier upon entry to a work-sharing construct, however there is an implied barrier at the end of a work sharing construct. 

There are three type of constructs:

# Do/for-loop

Shares iterations of a loop across the team. Represents a type of "data parallelism".

````
#pragma omp for [clause ...]  newline 
                schedule (type [,chunk]) 
                ordered
                private (list) 
                firstprivate (list) 
                lastprivate (list) 
                shared (list) 
                reduction (operator: list) 
                collapse (n) 
                nowait 

   for_loop

````

# Sections

Breaks work into separate, discrete sections. Each section is executed by a thread. Can be used to implement a type of "functional parallelism".

````
#pragma omp sections [clause ...]  newline 
                     private (list) 
                     firstprivate (list) 
                     lastprivate (list) 
                     reduction (operator: list) 
                     nowait
  {

  #pragma omp section   newline 

     structured_block

  #pragma omp section   newline 

     structured_block

  }
````

# Single

Serializes a section of code.

````
#pragma omp single [clause ...]  newline 
                   private (list) 
                   firstprivate (list) 
                   nowait

     structured_block
````


Supercomputers and HPC clusters

Components of a HPC system

HPC software stack

PBS - Portable Batch System

SLURM -Workload Manager

Parallel programming - OpenMP

Parallel programming - MPI

Parallel programming - GPU and CUDA

Conclusions and references

OpenMP - worksharing constructs