Data Warehouse Writeup

Mudrock University has asked you to help them to design their student grades data mart. Your requirements elicitation with the clients has determined that they are interested in answering questions such as:

• Number of students in each unit in each year
• Number of students in each unit offering (i.e., ICT394 in Semester 1 2017, is a different unit offering to ICT394 in Semester 2 2017) by year.
• Grade distributions of students in units by year
• Grade distributions of students in units by offering and year
• Grade distributions by Lecturer by unit and offering
• Grade distributions by Course
• Grade distributions by School

There are a number of data sources from which you will need to draw your data. Below are the sources and the tables within them that are/may be of interest to you. These are discussed below (Primary Key, Foreign Key):

Source 1: Course Handbook

The Course Handbook is a FileMaker Pro Database. It contains data regarding all courses, units and offerings of units that are offered by Mudrock University. A course is made up of units, and a unit will have at least one offering each year.:

COURSE (CourseCode, Version, CourseName, SchoolName)
UNIT (UnitCode, CourseCode, Version, UnitName)
UNIT_OFFERING (OfferingNumber, UnitCode, Year, TeachingPeriod)
OFFERING_COORDINATOR (StaffID, UnitOfferingNumber)

Source 2: Student Information System

The Student Information System has its data stored in a relational DBMS (Oracle) at present.

STUDENT (StudentID, StudentName, DateOfBirth)
ENROLMENT (EnrolNumber, StudentID, UnitOfferingNumber, Grade)

Source 3: Human Resources System

The HR System is a proprietary system that is owned by the HR Department.

STAFF_MEMBER(StaffNumber, StaffName, SchoolCode)
SCHOOL(SchoolCode, SchoolTitle)

What you have to do:

TASK 1: Discuss two (2) issues that may be problematic in the creation of the data warehouse that are apparent from the description above. For both, explain what you see as being the issue, why it is problematic, and what you will suggest needs to be done. This should take no more than two (2) pages in total.

TASK 2: Discuss what you see as being the most appropriate level of granularity for your data warehouse. Your discussion will need to explain why you have made this choice, and why the alternatives have been discarded. This should take not more than one (1) page.

TASK 3: Assuming that the issues you have raised in TASK 1 have been addressed to your satisfaction, design a Star Schema that will support the analyses as listed above.

TASK 4: Provide the SQL statements you would use to create the tables if you were to be implementing this design using Oracle.