Registration is closed.
The course will be taught on December 10 – 11 in virtual format.
Parallelware Trainer and Parallelware Analyzer are novel tools for the development of C/C++/Fortran parallel code for multicore CPUs and GPUs using OpenMP and OpenACC. Designed in collaboration with experts in parallel programming from High Performance Computing (HPC), they provide a systematic, more predictable approach that leverages parallel programming best practices, and enables the novice to write codes at the level of experts.
Appentra will present a practical course about vectorization and parallelization using Parallelware Trainer and Parallelware Analyzer that is a continuation of the “Introduction to parallelism” taught by CESGA. The course is intended to help new and existing programmers understand best practices for multicore CPU programming with OpenMP covering vectorization and shared-memory parallelization. The list of topics is shown in the table below.
During the course, the tools will collect anonymized usage of what tools and analyses are being invoked by the users. This is to help the further development work of the tools.
This event will be presented online using remote platforms.
Agenda
Day 1: Thursday, December 10th 2020
10:00-18:00
- Welcome and introductions
- Introduction to vectorization, parallelism and OpenMP for multicore CPUs
- Development of parallel codes with best practice recommendations
- Parallelware code patterns: compute patterns, memory patterns and flow patterns
- Parallelware tools: Trainer & Analyzer
- Parallelizing the calculation of Pi, MATMUL, HEAT.
- Lunch break
- Take-home exercises
- Hands-on with your own code
Day 2: Friday, December 11th 2020
9:00-14:00
- Office hours.
- Homework exercises demo, support, questions, FAQs using Parallelware tools
This course is aimed at researchers, professionals and university students interested in exploiting the resources offered by today’s computers to improve the performance of their applications.
Language: Spanish
Participation is free.
The course is limited to 40 attendees.
Professor: Dr. Manuel Arenaz, CEO of Appentra Solutions and professor at UDC.
Click here to read the Terms and Conditions.