Data warehouses based on the relational model has been a popular technology for many years, because they are very reliable due to their ACID-properties (Atomicity, Consistency, Isolation, and Durability). However, the new demands on databases today due to increasing amounts of data and data structures changing do mean that the relational model might not always be the optimal choice. NoSQL is the name of a group of databases that are less bound by schemas and are therefore more scalable and easier to make changes in. They are also adapted for massive parallel processing and are therefore suited for handling large amounts of data. Out of all of the NoSQL databases column-databases are the most like the relational model since it also consists of tables. This study has therefore converted a relational data warehouse based on a STAR-schema to a column-oriented-NoSQL-database and evaluated the implementation by comparing query-times between the relational data warehouse and the column-oriented-NoSQL-database. Scrambled economical data from a business in Sweden has been used to do the conversion and test it by asking a few usual queries. The results show that the mapping works but the query-time in the NoSQL-database is simnifically longer.