First Steps in Parallelization with OpenMP

Overview

We currently live in the era of multicore computers, where almost every computing machine has many CPUs. In order to benefit from this computational power we need to devise programs explicitly written for parallel machines. Nowadays there are several paradigms to write parallel code, and in this crash course we will focus on OpenMP. This is a minimally invasive open-source parallelization method that allows, in many cases, fast and straight-forward parallelization of currently working serial codes. OpenMP is meant for compiled code (eg C++ or Fortran) that run in a single computing node, but we will also visit some alternatives for Python, along with a short overview of parallelization over distributed memory machines via MPI.

Audience

This course is addressed to researchers who want to take their first steps in parallel computing. We will assume no previous knowledge of OpenMP (nor other parallelization techniques) from the participants.

Learning objectives

At the end of the course, the participants are expected to:

• understand and be able to implement basic parallelization methods with OpenMP

Prerequisites

Knowledge / competencies

Participants should be comfortable working in a Linux/UNIX environment and have some basic experience in programming. Some knowledge of C/C++ or Fortran is desirable, although not necessary.

Technical

You are required to bring your own wifi-enabled laptop. Windows users will need to install MobaXterm to enable ssh connections.

Application

Deadline for registration and cancellation is set is set to [21/11/2018]. Please note that participation to SIB courses is subject to our general conditions.

You will be informed by email of your registration confirmation.

Venue and Time

University of Basel, Klingelbergstrasse 61, Sky Lounge Biozentrum Annex, 13th Floor.

The course will start at 9:00 and end around 12:00. Precise information will be provided to the participants in due time.

Additional information

Coordination: Geoffrey Fucile, Anja Matthiä, and Rubén Cabezón

You are welcome to register to the SIB courses mailing list to be informed of all future courses and workshops, as well as all important deadlines using the form here.

For more information, please contact training@sib.swiss.