This page is under construction and updates are being made on a regular basis.

OpenMP (Open Multi-Processing) is a widely-used API that supports multi-platform shared memory multiprocessing programming in C, C++, and Fortran. It provides a set of compiler directives, library routines, and environment variables that allow developers to write parallel code for multi-core and multi-processor systems.

Programming CPUs using OpenMP

• Hello World example
• Parallel Construct
• Parallelize for loops
• Data Sharing Attribute Clauses
  • shared
  • private
  • firstprivate
  • lastprivate
• Master Construct
• Single Construct
• OpenMP Sections
• OpenMP Reductions
• OpenMP Scheduling
• Synchronization in OpenMP
  • Critical Sections
  • Atomic Operations
  • Barrier Synchronization
• OpenMP Tasking
• Other tutorials
  • Setting the number of threads in OpenMP using omp_set_num_threads

Programming GPUs using OpenMP – OpenMP Offload

• Hello World example
• Maximizing the GPU compute utilization
  • Teams Construct
  • Distribute Construct
• Optimizing the data transfers
  • Target data Construct
  • Target update Construct