...
CDAP captures metadata from many different sources—as sources, as well as those specified by a user—on user, on different entities and objects. The container model of CDAP provides for the seamless aggregation of a wide variety of machine-generated metadata that is automatically associated with datasets. This gives developers and data scientists flexibility when innovating and building solutions on Hadoop, without the worry of maintaining compliance and governance for every application.
CDAP metadata—consisting metadata, consisting of properties (a list of key-value pairs) or tags (a list of keys)—can , can be used to annotate artifacts, applications, programs, datasets, streams, views, and custom entities.
Using the CDAP Metadata HTTP RESTful API, you can set, retrieve, and delete these metadata annotations.
Metadata keys, values, and tags must conform to the CDAP supported characters, and are limited to 50 characters in length. The entire metadata object associated with a single entity is limited to 10K bytes in size.
...
Discovery and Lineage: Metadata can be used to tag different CDAP components so that they are easily discovered, identifiable, and managed. Lineage shows—for Lineage shows, for a specified time range—all range, all data access of the entity , and details of where that access originated from.
...