Database Reference
In-Depth Information
Figure 12. Multidimensional schema for the university data warehouse
Semester
Courses
Pk
Semester _Id
Pk
course_Id
Schools
Student
Sem_Name
Starting_Time
Ending_Time
Pk
School_Id
Pk
Suniv_Id
Course_Name
Course_Credit
Course_Desc
Semester_ID
School_Name
Univ_Name
Campus _Name
School_Location
Personal_No
Student_Name
Qualification
Location
Address
UniversityFact
Time
Pk
Program_Id
Pk
time_key
Pk
course_Id
Department
Pk
emp_Id
Date
Week
Month
Quarter
Year
Pk
time_key
Pk
dept_Id
Pk
Suniv_Id
Program
Dept_Name
School_ID
Dept_Desc
Ugrad_Offer
Pgrad_Offer
Countof_Studnt
Credits_Passed
Funds_acqird
Total _Funds
Expenses
Funds_Inhand
...
Pk
Program_Id
Dept_ID
Program_Name
Degree _Title
Admisn_Req
Degree _Req
Employees
Pk
emp_Id
Emp_Name
Qualification
Emplymnt%
Location
Address
Evolution operations:
There are two types
of changes in the example. a) Data changes, the
change which results in change in the data of the
DW; b) Schema changes, in which the schema of
the data warehouse is changed. In case of change
in degree requirements and increase in credits of
each course, data changes come into play, whereas
for the change in which a semester is divided into
two learning periods, this modification can be
handled by adding another level to in the dimen-
sion of programs (i.e. a level above course and
below semester). This type of change is called
schema change.
MVDW for the University case:
The problems
mentioned above motivate the need for using a
multiversion data warehouse for storing complete
and consistent information including overtime
changes in data.
According to the practical guidelines of prin-
ciple 1 (given in section 3.2), we label the root
version with V0.0. Due to changes in degree re-
quirements and changes in credits of courses the
DW will be inconsistent. Therefore, a new version
of the DW is derived from the root version after
time T1 and labeled as V1.0. Once a new version
is created, next data must be transferred to the new
version, labeled as V1.0. It is important to note that
a version V1.0 is created, but there is no change
in the schema of the data warehouse.
In order to handle the division of a semester into
two learning periods, after interval T2, it is required
to create a new version of the data warehouse as
well. The new version is labeled as V2.1 and is
derived from V1.0. The versioning graph for the
University DW is shown in Figure 13.
As shown in Figure 13, the root version of
university data warehouse is V0.0 and after time
T1 a new version (V1.0) is derived. Another ver-
sion (V2.1) is derived after time T2. In Figure 14,
we show the way in which schema sharing can
be modeled in accordance with the description
given in section 5.1.
Search WWH ::
Custom Search