Accelerating MATLAB Performance

290 Accelerating MATLAB

®

Performance

parfor idx = 1 : length(funList)

funcList{idx}(dataList{idx}); % this is run in parallel

end

When the computations are done, the workers pool should be closed or released:

delete(gcp)

Variables that appear in the body of a parfor-loop can be attributed to one of the classes

listed in the following table:

These variable classes are exemplied in the following code fragment:

c = pi; s = 0;

X = rand(1,100);

parfor k = 1 : 100

a = k; % a - temporary variable; k - loop variable

s = s + k; % s - reduction variable

if i <= c % c - broadcast variable

a = 3*a - 1;

end

Y(k) = X(k) + a; % Y - output ...