The design of
an ETL process aims at the production of a crucial deliverable: the mapping
of the attributes of the data sources to the attributes of the data warehouse
tables through the appropriate intermediate transformations. The production
of this deliverable involves several interviews that result in the revision and
redefinition of original assumptions and mappings; thus it is imperative that a
simple conceptual model should be employed in order to facilitate the smooth
redefinition and revision efforts and to serve as the means of communication
with the rest of the involved parties.
From our point of view, a conceptual model for ETL processes shall not be
another process/workflow model for the population of the data warehouse.
There are two basic reasons for this approach. First, in the conceptual model
for ETL processes, the focus is on documenting/formalizing the particularities
of the data sources with respect to the data warehouse and not in providing
a technical solution for the implementation of the process. Second, the ETL
conceptual model is constructed in the early stages of the data warehouse
project during which the time constraints of the project require a quick documentation
of the involved data stores and their relationships, rather than an
in-depth description of a composite workflow.
??? Logical.model:.In the logical perspective, we classify the design artifacts that
describe an abstraction of the workflow environment.
Pages:
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252