Efficient representation of your data

To get the best performance and resource usage, it is important that you organize your data in the DataSource of the UnstructGrid component as efficiently as possible. This will optimize the memory usage and also give the best rendering performance.

The main strategy is to avoid duplication of data and to store the data as compact as possible. All the objects in the data source are reference counted and can be used in many places.

Analysis with multiple states

If you only have rigid body movement of your geometry, please only define the geometry (elements and results) once and then use a DataResultTransformation result to specify one transformation matrix per part for each state.

If you have deformations, but the topology of the model is constant (no remeshing), please only define the geometry (elements and results) once and then use a DataResultDisplacement result to specify new nodes for each state (reusing the geometry).

If your model is remeshed for each state, you will have to specify the full geometry for each DataState.

Multiple geometries

An UnstructGridModel and it’s DataStates can have multiple DataGeometry per state. This allows combinations of the three options above to further reduce the resource usage. So if some parts of your model are static while others are remeshed, CEETRON Envision handles this efficiently if you split your model into two geometries and provide the data for each geometry in the most optimal way.