so for example, if you want to do

    M = [1 0 0 0
         0 1 0 0
         0 0 1 0
         0 0 0 1]
    v1 = [1,2,3,4] (column vector)
    v2 = [4,5,6,7] (column vector)

you can either do (M * v1, M * v2) on two cores as

    r1a = [1 0 0 0
           0 1 0 0] * v1  (core 1)
    r1b = [0 0 1 0
           0 0 0 1] * v1  (core 2)
    r2a = [1 0 0 0
           0 1 0 0] * v2  (core 1)
    r2b = [0 0 1 0
           0 0 0 1] * v2  (core 2)

then

    r1 = flatten([r1a, r1b])
    r2 = flatten([r2a, r2b])

OR, you could just have the whole model in each core:

    r1 = M * v1 (core 1)
    r2 = M * v2 (core 2)