Solve Sparse, Complex Ax=b on Distributed Cluster

I would like to know if it is possible to solve a sparse, complex Ax=b equation with Parallel Computing Toolbox / Distributed Computing Toolbox. It seems that this is not possible with mldivide. I tried using mldivide on a sparse co-distributed array, but got an error: "Error using \ (line X) Sparse input arguments are not supported.".

PCG is not a good option because my matrix is not positive definite.

Any thoughts regarding things I should try?

[Remington Reid]

I have also been searching for a good way to solve this problem. If I find something good I will post it.