The Luxembourg SuperComputing Competence Center is hosting an online introduction to OpenMP programming (for multicore shared memory) course for half a day. The first part will be dedicated to theory, and the second will focus on hands-on challenges on the MeluXina supercomputer.
Both current or prospective users of large hybrid CPU/GPU clusters and supercomputers, who might develop or parallelise their scientific computing applications using multicore CPUs, are encouraged to participate!
What will you learn and how?
Participants from this course will learn Multicore (shared memory) CPU programming using the OpenMP programming model, such as parallel region, environmental routines, and data sharing. Furthermore, understanding the multicore shared memory architecture and how parallel threads blocks are used to parallelise the computational task. Since we deal with multicores and parallel threads, proper parallel work sharing and the synchronisation of the parallel calls are to be studied in detail. Finally, participants will also learn to use the OpenMP programming model to accelerate linear algebra (routines) and iterative solvers on the Multicore CPU. Participants will learn theories first and implement the OpenMP programming model with mentors' guidance later in the hands-on tutorial part.
After this course, participants will be able to:
Understanding the shared memory architecture
- Unified Memory Access (UMA) and Non-Unified Memory Access (NUMA)
- Hybrid distributed shared memory architecture
- Parallel region
- Environment routines
- Data sharing
- Work sharing
- Synchronisation constructs
- Single Instruction Multiple Data (SIMD) directive
- Iterative solvers from science and engineering
- Vector multiplication, vector addition, etc.
Priority will be given to users with good experience with C/C++ and/or FORTRAN. No prior parallel programming experience is needed.