Broadcasting a Better Value in MPI

Question

I want to write a small program in MPI (Java implementation) A variable x (double variable) is declared. Threads try to modify the variable (let's say a random modification). When a thread i finds a new value of X which is smaller than the older one, a broadcasting to other threads is done so that they can update the value of their variable X

I have looked at the Bcast function in MPI ... but in all examples it was called by all threads whether the variable is modified or not.

That's right, that's how broadcasts work - all the processes in a communicator take part. Until the processes exchange messages one process doesn't know that the value of a variable on another process has changed. You might want to look at other collectives such as mpi_gather, mpi_scatter or mpi_allgather. There are others too. — High Performance Mark

Hristo Iliev Hristo Iliev · Accepted Answer · 2014-09-23T18:10:58

This is one of those scenarios that are quite easy to implement in a multithreaded environment (e.g. OpenMP or Java threads) and very hard to impossible to implement efficiently in MPI. The usual approach is to refactor your algorithm in such a way that the best value could be communicated every N steps (with N possibly equal to 1, but that could be very inefficient due to the communication overhead) and then use Intracomm.Allreduce with the reduce operation set to MPI.MIN. Each process provides its own minimum value and the reduction returns the global minimum. If you would also like to know the rank of the process that holds the global minimum value, MPI.MINLOC should be used instead.

If you are trying to implement parallel genetic optimisation, there are some C++ libraries that might give you an inspiration.

Broadcasting a Better Value in MPI

1 Answers