For example, Strange (2002) mentions the development of a data warehouse realized
in the Fortune 500 financial institution. This development included the support of
the data warehouse for applications to perform customer retention analysis, bank
loan risk management, customer contact history, and many other applications. There
were 100 people on the data warehouse team (approximately 8.5% of the overall
IT staff)??”55 from ETL, four database administrators, four architects, four systems
administrators, nine BI competency center workers (assisting end users), five report
writers, nine managers, and nine hardware, operating system, and operations support
staff members. These 55 individuals were responsible for building and maintaining
the ETL process, which includes 46 different source systems. Responsibilities
include updates of data marts on a weekly and monthly basis. This does not include
staff from operations to support the execution of the ETL processes. They used a
large parallel server platform that is consisted of multiple silver nodes (four processors
per node) and four terabytes or more of disk storage, at an acquisition cost
over three years of US$5 million. The cost of the ETL tool used was US$1 million,
excluding the yearly maintenance and support costs.
Moreover, these processes are important for the correctness, completeness, and freshness
of data warehouse contents, since not only do they facilitate the population of
the warehouse with up-to-date data, but they are also responsible for homogenizing
their structure and blocking the propagation of erroneous or inconsistent entries.
Pages:
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242