Split a matrix on parallel computing with spmd

As I can split a vector 4x18 in four workers each with a 1x18 table, selecting the vector for each worker in parallel computing with spmd.

[Walter Roberson]

That would sound like something more natural for parfor ?

[Cèsar]

Hi, Walter How can i do with parfor.

Answer 1

[Edric Ellis]

You could do that using distributed arrays. For example

    spmd
        d = codistributed(rand(4, 18), codistributor1d(1))
    end

will result in each worker storing a 1x18 portion of d.

Answer 2

[Cèsar]

Thank you so much, it works really good. But if i want to distributed the array in a some especific way in to the workers, for example if i had 4 workers and a array of 5x18 and i want in the first worker an array of 1x18, in the second 1x18, the third 2x18 and in the forth 1x18,how could i do, thank you so much for your answers.