Collective communication using broadcast
During the development of a parallel code, we often find ourselves in the situation where we have to share between multiple processes the value of a certain variable at runtime or certain operations on variables that each process provides (presumably with different values).
To resolve this type of situations, the communication trees are used (for example the process 0 sends data to the processes 1 and 2, which respectively will take care of sending them to the processes 3, 4, 5, and 6, and so on).
Instead, MPI libraries provide functions ideal for the exchange of information or the use of multiple processes that are clearly optimized for the machine in which they are performed.
Broadcasting data from process ...
Get Python Parallel Programming Cookbook now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
