As the environment in advanced technique of smart-phone and wireless communication has been developed, mobile computing has been served based on embedded system in order to carry out specific function demanded by user. In accordance with increase of the application by the use of locational real-time data such as traffic information, the method to efficiently manage massive spatial data is required. Therefore, the object of this paper is to suggest the storage model regarding binary stream format of spatial data based on relational indexing, extending open source embedded DBMS (SQLight). First of all, designing physical storage model for relational indexing, we define binary stream for geometry of spatial data according to WKB (Well-Known Binary), compression algorithm (Precision, Bytepacking) and compression library (Zlib). Finally, we respectively compare the results of record set and execution time retrieved from query processing. As an outcome, suggested streaming data scheme based on relational indexing will be utilized in efficient query processing for unformatted spatial data.