We
have stressed the fact that there is not a unified approach that concretely deals with
the modeling and the optimization of ETL workflows.
0 Simitsis, Vassiliadis, Skiadopoulos, & Sellis
Copyright ?© 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of
Idea Group Inc. is prohibited.
Additionally, we have indicated the main challenges and problems concerning the
manufacturing of ETL workflows. In all the phases of an ETL process (extraction
and transportation, transformation and cleaning, and loading) individual issues
arise, making data warehouse refreshment a very troublesome task. In general, ETL
workflows are characterized as quite complex, costly, critical, and important for the
success of a data warehouse project. The key factors underlying the main problems
of ETL workflows are vastness of the data volumes, quality problems, performance,
evolution of the sources, and the data warehouse.
Moreover, we have presented a modeling approach for the construction of ETL
workflows, which is based on the life cycle of the ETL processes. This life cycle
consist of four phases: reverse engineering and requirements collection, logical
design, tuning and physical design, and software construction. As a result, in order
to construct an efficient, robust, and evolvable ETL workflow, we have to deal with
all the phases of the life cycle of a data warehouse.
Pages:
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267