DAAL does not provide matrix multiplication functions, although this is a fundamental operation internally used by many other algorithms. Can you please explain how you want to use distributed matrix multiplication? For example, maybe you need it in the implementation of an algorithm? We will see if DAAL has already had the solution.
Meanwhile, if you are interested in making the matrix multiplication in MLlib run faster, you can try to rebuild MLlib by using Intel MKL as its underlying math lib.