Consider an Engineering database that maintains three tables: EMP (ENO, NAME, TITLE, SAL) PROJ (PNO, PNAME, BUDGET, LOC) ASG (PNO, ENO, RESP, DUR) EMP stores employee information, including the name, title and salary of the employee. PROJ stores project information, including the name, budget and location of the project. ASG keeps information about the assignment between projects and employees. RESP is the responsibility of the employee, while DUR is the duration that the employee works on the project. ENO is the employee id. PNO is the project id. The keys are underlined. The following predicates appear in the majority of queries. LOC = New York, LOC = Montreal, LOC = Paris SAL > 5000, SAL 5000 a) Write down the fragments if we perform primary fragmentation on EMP and PROJ, given the predicates above. Is this kind of fragmentation a good one? Why? (Hint: Analyze the desired properties of fragmentation.) b) Write down the fragments if we perform derived horizontal fragmentation on ASG based on the fragments of PROJ in a). Is this kind of fragmentation a good one? Why? (Hint: Analyze the desired properties of fragmentation.)