Where does a DBMS store persistent data

Where does a DBMS store persistent data? How does it bring data into main memory for processing?
What DBMS component reads and writes data from main memory, and what is the unit of I/O? (Section 8.1)

2.

What is a file organization? What is an index? What is the relationship between files and indexes? Can
we have several indexes on a single file of records? Can an index itself store data records (i.e., act as a
file)? (Section 8.2)

3.
4.

What is the search key for an index? What is a data entry in an index? (Section 8.2)
What is a clustered index? What is a primary index? How many clustered indexes can you build on a
file? How many unclustered indexes can you build? (Section 8.2.1)

5.

How is data organized in a hash­based index? When would you use a hash­based index? (Section
8.3.1)

6.

How is data organized in a tree­based index? When would you use a tree­based index? (Section
8.3.2)

7.

Consider the following operations: scans, equality and range selections, inserts, and deletes, and the
following file organizations: heap files, sorted files, clustered files, heap files with an unclustered tree index on
the search key, and heap files with an unclustered hash index. Which file organization is best suited for each
operation? (Section 8.4)

8.

What are the main contributors to the cost of database operations? Discuss a simple cost model that
reflects this. (Section 8.4.1)

9.

How does the expected workload influence physical database design decisions such as what indexes
to build? Why is the choice of indexes a central aspect of physical database design? (Section 8.5)

10.

What issues are considered in using clustered indexes? What is an index­only evaluation method?
What is its primary advantage? (Section 8.5.2)