Endeca Server - Data (Model|Structure)
About
The data structure in the Oracle Endeca Server consists of:
- records. Records are the fundamental units of data.
- and attributes. An attribute is the basic unit of a record.
The Oracle Endeca Server:
- stores each column of information on a combination of disk and memory in two sorts orders (by value and by a universal record key identifier).
- contains a tree-structured index in memory to speed the retrieval of information.
Articles Related
Characteristics
- No meta-data predefinition:
- Table free: only record are loaded where its size can vary
- faceted data model: each record is made of a series of name-value pairs (facets)
- Jagged records: the record size may vary between set and each name-value pairs is not obligatory defined for each set.
- Structured and unstructured content
- Fully Denormalized, all attributes are onto the record
Advantages
- As the as the “Schema” follows the data, new (data|information|business process) can be:
- quickly added by loading new records
- done at any time
- Disparate data: structured and unstructured where unstructured data can be analysed, enhanced and serve as filter for structured data
Data Structure
Records
Almost all information including raw data (data record) and the data schema (system records), is represented by records.
| Record type | Description |
|---|---|
| Data records | Input Data that you want to explore. |
| System records | System records store configuration information. Changing these records controls the behaviour of your data records. |
| Primordial records | Primordial records are created automatically and used internally by the Oracle Endeca Server. |
Schema
A collection of records with Managed attributes:
- forms a schema for your records.
- is known as system records.