With the explosion of data, new tools become available. One of the most popular tools is the open source Apache Hadoop Framework. In this thesis, the MapReduce program model is examined, together with the Hadoop framework any open matters in the field of the databases that is called to cover. It is examined cases that the different architectures of databases systems overlay each other or each one of them is requested to fill in the open issues of the other architecture. As well it is examining the Hadoop framework is responding in demands of structured data and more specifically in cases of geospatial values, although the Hadoop Framework is oriented to unstructured data of text type. Furthermore it is evaluating the advantages of Hadoop, which could be fully used by spatial system Hermes, which is created by the Information Dept. of Piraeus University. The first scenario, of which operational features was examined is the co-operation of Hadoop with Hadoop DB which is basically reflects the co-operation of a shared-nothing architecture with a system RDBMS (Post GIS). This solution was finally judged as none “usable”, because as explained below it fits to the characteristics of shared-nothing architecture, only at the minimum. The second scenario, which is finally our proposition, is the access to spatial data, which are stored in a HDFS system by using HIVE and extended tools that are provided by third parties and are customized in spatial querying.