|
|
|
Data Warehouse
The University implemented a data warehouse solution in 1999
essentially to overcome the difficulties of extracting ad hoc queries from
a custom built student information system.
The data warehouse empowers
users to extract student data independently of specialist IT resources and
therefore improves query turnaround time, data integrity and the quality
of information provided to management. The project reflects
the transition from the traditional model of information governance to
information democracy. Opportunities to transform data into
information into knowledge become valuable commodities to the
organisation. Importantly, the University became focussed on business
requirements supported by IT rather than IT driven processes.
Implementation of the data warehouse
Following a feasibility study, the University decided to adopt
an institution wide standard to data extraction, reporting and analysis
for effective communication of information. Formal acceptance testing
procedures found that the
Business Objects tool was able to fulfil the
University's requirement to support an integrated decision support
solution.
A minimalist approach in a phased
implementation was adopted for two important reasons;
-
availability of resources at a time when
other resource intensive projects such as the implementation of an
Enterprise Resource Planning system (SAP) were undertaken, and
-
pressing demands from users unable to
effectively perform ad hoc queries on student data.
The project proved successful primarily due
to its rapid implementation, consultative and user driven approach and
ability to access data via a versatile and user friendly decision support
tool.
Future developments for the data warehouse
will include data sources such as research, finance and human resources.
What is a data warehouse
A data warehouse is a collection of integrated databases
specifically structured for querying, analysis and reporting - unlike production
databases that are optimised for transactional processing (for example,
the student information system).
Data
The data stored in the warehouse is extracted from the student
information system. To isolate users from the technical aspects of the
database, designers create universes to map to data in the database in
everyday business terms. Universes consist of logical groups of classes
and objects that reflect different operations of the organisation. For
example, from the student information system, higher degree, fee
distribution and enrolment universes have been created.
Data in the warehouse is refreshed daily
providing access to student data from the previous day. The refresh occurs
overnight between 10pm and 8.30am to ensure minimal interruption to end
users.
Access
The interface to the data warehouse is the Business Objects
reporting tool. As part of a quality assurance program,
training
is mandatory before access to the data warehouse is provided. Users will
gain an understanding of the functionality of Business Objects and gain
insight into the underlying databases specific to the University's data
warehouse.
Article on Data Warehouse (published in the Journal of
Institutional Research produced by the Australasian Association for
Institutional Research):
Mukerjee,
S. (2002) Data Warehousing: piloting a minimalist approach, Journal of
Institutional Research, vol 11, no. 1, pp 71-77.
|
|
