MPI collective communication can be divided into three subsets:

Synchronization

• Barrier synchronization

Data Movement

• Broadcast from one member to all other members
• Gather data from an array spread across processes into one array
• Scatter data from one member to all members
• All-to-all exchange of data

Global Computation

• Global reduction (e.g., sum, min of distributed data elements)
• Scan across all members of a communicator

There is a separate topic in this roadmap for each of these three types of collective communication.