MAGIC (MAG-netic Intelligent Compound) is a solidified magnetic ferrofluid (MF) containing both magnetic particles (MPs) and abrasive particles (APs, nonmagnetic) of micron size. The distribution of APs in MAGIC can be controlled by applying a magnetic field during cooling process of MAGIC fluid. Due to its lower simulation speed, traditional Stokesian Dynamics method is limited to a small amount of particles. To speed up the simulation and simulate more particles, cluster-based Stokesian Dynamics (CSD) method is applied in the present study. Some attempts about parallel computing of CSD and other parts of our simulation algorithms are also carried out and some results of time comparison on multi-core computers are summarized. Based on the above method, effects of magnetic field, size and concentration of particles on their final distributions are examined, which shows that strong magnetic field, high concentration and small-size particles can induce more uniform distributions of APs in MAGIC fluid, thus providing theoretical basis for manufacturing of MAGIC polishing stone.