Epidemic spreading is still an attractive topic for public attention because of the regular occurrence of pandemics throughout history. Authorities collect health statistics based on geographic divisions and make it open for everyone to create potentially impacted tools. In this study, we give an approach of using distributed process systems for epidemic spreading simulation with the segmentation of geographic divisions. A cellular automata model is defined on this irregular cell space with the initial conditions acquired from Open Data repositories. The spreading process is performed by local exchanges in an adjacent neighborhood. Open data repositories offer multiple parameters that can be used to approximate local behaviors inside automata specifications. Experiments are conducted on Covid-19 spreading simulation and data in a one-year period is analyzed to deduce the transition function. Experiment results show that the epidemic propagation trend is caught although the simulated incidence rates are generally lower than the real incidence rates collected. The practical interest is to understand epidemic spreading in time and space. Principles can be reproduced in a number of situations provided that accurate geographic segmentation and related data are available.