Get an overview of these long-lived parallel programming workhorses, including their strengths and the types of problems each is best suited for.
Two decades is a millennium in technology years. And yet … Message Passing Interface (MPI), Open Multi-Processing* (OpenMP*) and Intel® Threading Building Blocks (Intel® TBB) have made the cut, helping the global developer community parallelize code for 25, 20 and 11 years, respectively. Pretty impressive. And they remain popular largely because they’re based on open source and standards-driven implementations, and they offer intuitive approaches to parallelism. Join us in this webinar where we’ll:
- Explore the strengths of each parallel model
- Discuss the types of problems for which each is best suited
- Describe the latest features of each (spoiler alert :: the latest OpenMP specs provide directive-based approaches to accelerator offload and vectorization, and Intel TBB flow graph classes and functions allow easy expression of unstructured parallelism, dependency graphs, and data flow algorithms)
Cover the programming tools available to develop, debug and tune MPI, OpenMP and Intel TBB programs. If you haven’t yet, be sure to download both Intel® MPI Library and Intel TBB—part of the free Intel® Performance Libraries.