Name: 2019 Oil & Gas Conference Talk
Start: 2019-3-5T13-13-45-00
End: 2019-3-5T14-14-15-00
Location: Rice University 6500 Main Street Houston, TX , USA

A hands-on introduction to using Parallelware Trainer

Using Parallelware Trainer to speed up your understanding of OpenMP and how to use OpenMP in real applications.

Details

High Performance Computing (HPC) is fundamental in solving scientific problems, from protein folder to developing nuclear fusion reactors, however, it is often viewed as complex and difficult to use. Traditional training in HPC software development, while a great foundation for the theory, often leaves attendees with a knowledge gap on how to tackle real-applications rather than the simple examples given in training.

This 3 hour course will demonstrate how to use Parallelware Trainer and to simplify your learning of OpenMP, and how to quickly and effectively use OpenMP in real applications. Find out in the practical sessions how to easily take a real application from sequential code to OpenMP enabled software in just a few hours. By using Parallelware Trainer you will also be able to minimise and nearly eliminate the introduction of syntax errors and race conditions as you are guided through the parallelisation process.

This foundation will give you the ability to appreciate how to use HPC in your field and also equip you with the tools to start making effective use of HPC facilities yourself.

On completion of the course, we expect attendees will be in a position to use Parellelware Trainer for continued self-learning of OpenMP and even start using OpenACC for accelerators. Attendees will also have all the tools needed to start developing their own OpenMP enabled applications.

The course is delivered using a mixture of lectures and practical sessions and has a very practical focus.

Intended learning outcomes

On completion of this course students should be able to:

Use Parallelware Trainer on their own applications.
Understand your code in terms of parallel patterns by using Parallelware Trainer to quickly identify the patterns and facilitate the introduction of parallelism.
Understand what variables need to be protected when introducing OpenMP directives.
Maximise limited software development time for improved code correctness and performance.

Course Materials

Slides (PDF)
Hand out Parallelware Trainer help sheet (PDF)
Worksheet (PDF)
Hand out profiling help sheet (PDF)
Parallelware Trainer manual (PDF)

Questions?

If you have any questions please contact us.

Upcoming courses

january, 2020

No Events

Finished courses

march 2019

tue05mar13:45tue14:152019 Oil & Gas Conference TalkDemocratising Access to HPC13:45 – 14:15 Rice University 6500 Main Street Houston, TX , USA

may 2019

wed22mayAll Daythu23PATC: Heterogeneous Programming on GPUs with MPI & OmpSsPRACE Training: Heterogeneous Programming on GPUs with MPI & OmpSs(All Day) BSC Supercomputing Center, Barcelona , Spain

june 2019

thu06jun09:00thu17:00NERSC Parallelware Tool WorkshopOne-day training event on using the ParallelwareTrainer tool to learn to add OpenMP or OpenACC directives to your code.09:00 – 17:00 Berkeley, CA. Building 59, Room 4102 (CRT Building).

october 2019

thu17oct08:15thu17:00NERSC Parallelware Tool WorkshopUsing the Parallelware Trainer tool to learn to add OpenMP or OpenACC directives to your code.08:15 – 17:00 Berkeley, CA. Building 59, Room 4102 (CRT Building).

thu17oct14:00thu17:30BSC PATC Parallel Programming WorkshopParallel Programming with Parallelware14:00 – 17:30 BSC Supercomputing Center, Barcelona , Spain

thu31octfri01novOpenPOWER Summit EU 2019Tutorial Session: GPU Programming Using Patterns for Parallelization 17:30 – (november 1) 13:00 Marriot Lyon Cite Internationale, France.

Event Details

Tutorial Session: GPU Programming Using Patterns for Parallelization
Dr. Manuel Arenaz & Dr. Javier Novo, Appentra Solutions

Tutorial Session Pre-Requisite: Thursday, October 31 • 17:30 – 18:10. Read more: https://sched.co/WdM9
Tutorial Session: Friday, November 1 • 09:55 – 13:00. Read more: https://sched.co/Vha1

As HPC continues to move towards a model of multicore and accelerator programming, a detailed understanding of shared-memory models and how best to use accelerators has never been more important. OpenMP is the de facto standard for writing multithreaded code to take advantage of shared memory platforms, but to make optimal use of it can be incredibly complex. OpenACC is considered a forerunner and simpler to use, which is why it is being successfully used today for the acceleration of real applications.

This course will give a holistic understanding of multicore and GPU programming using a novel productivity-oriented approach based on the decomposition of real applications into components for parallelization. The workshop will cover:
– Introduction to multicore and GPU programming.
– Code patterns for parallelization.
– OpenMP and OpenACC.
– Multithreading, offloading and tasking.”

Watch the event in video: https://www.appentra.com/training/courses/event-training-courses/#openpower

Time

October 31 (Thursday) 17:30 – November 1 (Friday) 13:00

Location

Marriot Lyon Cite Internationale, France.

Learn More

Calendar GoogleCal

november 2019

thu31octfri01novOpenPOWER Summit EU 2019Tutorial Session: GPU Programming Using Patterns for Parallelization (october 31) 17:30 – (november 1) 13:00 Marriot Lyon Cite Internationale, France.