Historical Relational in HRDM
The HRDM model, presented in [CC87] and building on the work reportde in [Cli82b]
(the HDBM model) and [CT85], was one of the early historical database models presented in
the literature.HRDM is
- A temporally grouped model
- Providing for time-varying data and temporal grouping.
- Provided, to a limited extent, for time-varying schemas
within the same framework.
- An inhomogeneous or heterogeneous model.
In order to address these temporal issues, we use the lifespan notion introduced in [CT85]
. The lifespan is valid time of a data object, we could use [t1, t2] to repersent it and
use mathematic "SET ALGEBRA" to operate them.
When we want to handle temporal data. we define
D = {D1,D2,......Dn}
T = universal time lifespan
TD = {TD1, TD2, TD3,...TDn}, where for each i , TDi = {fi|fi:T->Di} is the set of all
partial function from T into the value domain Di.
TT = {g|g:T->T} is the set of all partial functions from T into itself.
U={A1, A2,......,An} be a (universal) set of attributes.
All attributes in the historical relational data model are defined iver sets of parial
temporal function.
HD = (TD or {TT}) = { TT, TD1, TD2....TDn }
is the set of all historical domains.
Therefore, all temporal data could represent a tuple t on scheme R like, t = [v,l],
and calculate by some operator.
- Set-Theoretic Operation
- Project
- Select
- Time-slice
- When
- Join