OpenMP® is an API that multi-threads C, C++ and Fortran code for parallel processing on multi-core CPUs and Intel Xeon Phi Coprocessors. It is a simple way to take advantage of all of the cores in modern computers, providing a portable and scalable model that is suitable for range of platforms from the standard desktop computer to the supercomputer.
This two day OpenMP course is designed to teach programmers how to develop high performance parallel applications.
The course content and hands-on exercises have been developed from Acceleware’s extensive experience of using OpenMP both for our commercial seismic depth imaging libraries and from our outbound consulting services provided to the oil and gas, financial, computer aided engineering, defence, medical and entertainment industries. Some examples of OpenMP projects we have completed include: Seismic data filtering, Kirchoff depth migration, reverse time migration, Computed Tomography (CT) back filtering imaging reconstruction and option pricing for financial organisations.
Learn from the best
The courses are taught by Acceleware programmers who bring real world experience into the classroom. To date Acceleware has delivered over 100 courses across four continents, teaching hundreds of programmers how to design high performance applications.
2 Day Course outline
- Course introduction
- Introduction to OpenMP
- OpenMP programming model
- Synchronization and Library Routines
- Hands-on-exercises: Launching a multi-threaded application and arithmetic reduction
- Work-sharing directives & scheduling clauses
- Debugging and profiling techniques
- Environment variables and MPI+OpenMP
- Hands-on-exercises:Matrix multiply, debugging and filtering case study
- Small class sizes to maximize learning
- Hands-on exercises developed by our experts
- Individual laptops for student use
- Choice of Windows or Linux operating system
- Printed manual of all lectures
- Electronic copy of lab exercises
- Certificate of completion