Discussed is how we can build a data-assimilative model by augmenting a forecast model by data assimilation functionality for efficient ensemble data assimilation. The implementation strategy uses a direct connection between a coupled simulation model and ensemble data assimilation software provided by the open-source Parallel Data Assimilation Framework (PDAF, http://pdaf.awi.de), which also provides fully-implemented and parallelized ensemble filters. The combination of a model with PDAF yields a data assimilation program with high flexibility and parallel scalability with only small changes to the model. The direct connection is obtained by first extending the source code of the coupled model so that it is able to run an ensemble of model states. In addition, a filtering step is added using a combination of in-memory access and parallel communication to create an online-coupled ensemble assimilation program. The direct connection avoids the common need to stop and restart a whole forecast model to perform the assimilation of observations in the analysis step of ensemble-based filter methods like ensemble Kalman or particle filters. Instead, the analysis step is performed in between time steps and is independent of the actual model coupler. This strategy can be applied with forced uncoupled models or coupled Earth system models, where it even allows for cross-domain data assimilation. The structure, features and performance of the data assimilation systems is discussed on the example of the ocean circulation models MITgcm and NEMO.