MPI Performance Guidelines

Welcome to the MPI Performance Guidelines site! The creation of this site was the result of a BSSw Fellowship. Please check out the initial set of guides and their companion videos.

• Avoiding Unintended Synchronization - [video]
• Ensuring Progress for MPI Nonblocking Operations - [video]
• Minimizing Thread Contention Using Communication Objects - [video]
• Dynamic Sparse Data Exchange with MPI_IBARRIER - [video]

Want to contribute? See here for more information.