In the era of big data, data is of great value as an essential factor of production. It is of great significance to implement its analysis, mining and utilization of large-scale data via data sharing. However, due to the heterogeneous dispersion of data and increasingly rigorous privacy protection regulations, data owners can not arbitrarily share data. This dilemma turns data owners into data silos. Data Federation calculate collaborative query while preserving the privacy of data silos. This study implements a multi-party secure relational data federation system. The system is designed based on the idea of federated computation that “data stays, computation moves”. Its adaptation interface of the system is different kinds of relational database adaptation, which can shield the data heterogeneity of multiple data owners. The system implements the multi-party security basic calculator library based on secret sharing, and the calculator realizes the optimization of the result reconstruction process. On this basis, it supports the query operations such as sum, average, maximum, equi-join and theta-join. Making full use of the multi-party properties to reduce the data interaction among data owners, the proposed system reduces the security computation overhead, so as to effectively support efficient data sharing. Finally, the experiment is carried out on the benchmark data set TPC-H. The experimental results show that the proposed system can support more data owners’ participation and has higher execution efficiency than current data federation systems such as SMCQL and Conclave by at most 3.75 times.