Your data can initially be in various formats (xls, txt, rtf, doc, csv, odbc, soap, etc.)
To be published and useful as open data, it need to become "usable" structured and machine readable as there will be the Data Consumer - user, who will search, filter or classify your data, or even develop some new application working with this data.
This activity implies the transformation of your data into open non-proprietary formats.
Therefore, on the output, your datasets will be (following your choice) provided to data consumers:
in the dumps (CSV for tabular data, RDF for linked data),
published via API (SPARQL Endpoint for RDF data, REST API for tabular data).
Data administrators/publishers may also configure some specific settings per each publication option: to tweak dump generation process (like which RDF serialization to use: Turtle, XML, etc.), to select which resources (tables, graphs) associated with the transformed dataset (and stored in ODN/Storage) should be published - made available to data consumers, etc.
Between your initial state of the art (mostly the "not too structured" dump of documents, tables, records,...) and the final form of searchable, well ordered, user friendly datafiles, there are some well defined procedures to be realised in ODN.
To publish your data as Open Datasets you will then follow publication path
If you already know what to do, follow to section How to ...?.