Neptune's Data Model
Neptune is a sparse, distributed, persistent multidimensional sorted
map(from Bigtable's data model).
- Table: All datas are stored in logical data
storage called "Table". Table is identified by table name and a table
name must be unique in a neptune cluster.
- Column: A table has multiple columns. A columns has same
meaning RDBMS's column. Column must be defined when table is created or
can be added with "add column" instrunction.
- Row: A table consists of numerous rows.
- Row.Key: Each row has unique identifier,
"rowkey". Neptune automatically creates rowkey, while user
explicitly specifies row key column in a relational database. Data is
sorted by Row.Key. Row-Key is String type.
- Cell: Cell is a intersection of row and column. Cell can
have many key-value pairs(Cell.Key-Cell,Value).
- Cell.Key: Datas in a cell identified by Cell.Key and sored
by Cell.Key. Cell.Key is String type.
- Cell.Value: A Cell.Value has multiple versions and
sorted by timestamp in descending order. Each version stores byte
array.
