All data in TileDB is modeled either as a dense or a sparse array. This of course includes dataframes. In other words, dataframes are arrays, and, as such, all functionality described in Arrays applies to dataframes as well. However, we have implemented special functionality for creating, ingesting and reading dataframes to make usage more natural to users familiar with dataframes. For example, we provide SQL support via MariaDB, integration with Pandas and Arrow Table in Python, etc.