I am trying to run multiple instances of a function using 10 workers in a parallel pool. One instance of a function takes around 20 hours to run that is why I am using parallel computing so that I could get 10 results in 20 hours. But, after a few hours of starting the parallel computing, the number of workers automatically get reduced to 2. I am not sure what is happening. Here is the code for parallel computing that I am using:
clear all;
clc;
close('all')
numOfIterations = 60;
delete(gcp('nocreate'));
p = parpool(10);
% To request multiple evaluations, use a loop.
for idx = 1:numOfIterations
f(idx) = parfeval(p,@parallel_computing_distributed_fun,1,(idx+1043)); % Square size determined by idx
end
% Collect the results as they become available.
counter = numOfIterations;
for idx1 = 1:numOfIterations
%fetchNext blocks until next results are available.
flag = fetchNext(f);
disp('Completed the fetch of')
disp(idx1 + 1043)
disp(clock)
end
The pool does start with 10 workers. Here is output on the command window:
Parallel pool using the 'local' profile is shutting down.
Starting parallel pool (parpool) using the 'local' profile ...
Connected to the parallel pool (number of workers: 10).
But after a while the number of workers gets reduced to 2. Can someone please help me in finding out why the number of workers are getting reduced and how to fix it?
