Abstract
MapReduce frameworks, e.g., Hadoop, have been used extensively in different applications that include machine learning, and spatial processing. In meantime, huge volumes of spatio-temporal trajectory data are coming from different sources over sometime, raised the demand to exploit the efficiency of Hadoop, coupled with the flexibility of the MapReduce framework, in trajectory data processing. This work describes Summit; a full-fledged MapReduce framework with native support for trajectory data.