1) What is a log-file group?

Ans: A log-file group is a set of log files that is used to store system redo information. By having a group of log files, Oracle will automatically mirror redo information to the log files in the groups.

2) How many I/Os per second per disk drive can a standard disk drive service?

Ans: You should not run more than 60-70 I/Os per disk drive on a standard SCSI disk drive. Pushing more I/Os than this can cause disk latencies to increase beyond recommended limits.

3) What is an SID?

Ans: The SID is the system identifier. This environment variable is used to determine which database to connect to.

4) How many databases can you create with one SID?

Ans: Each SID uniquely identifies one database; therefore you can only create one database per SID.

5) How many databases can you create on one system?

Ans: There is no firm limit to the number of databases that can be created on one system. The system resources that each database consumes will be the limiting factor.

6) What is a tablespace used for?

Ans: The tablespace is used to create schema objects. Tables, indexes, views, and clusters all are created within tablespaces.

7) What is an extent?

Ans: An extent is the unit of space that a schema object allocates when it is created or grows. As a table increases in size and fills up all its space, another extent is allocated. An extent is a contiguous group of blocks.

8) What states can a tablespace be in?

Ans: tablespace can be online or offline.

9) Name the four types of segments.

Ans: The four types of segments are data, index, rollback, and temporary.

10) What DBMS object is used for recovery operations?

Ans: The redo log files and the archive log files are used for database recovery.

11) What DBMS object is used for read consistency?

Ans: The rollback segments are used for read consistency.

12) What is read consistency?

Ans: Read consistency allows a long-running transaction to always obtain the same data within the query.

13) How many redo log files do you need?

Ans: Oracle requires at least two redo log files. This allows the log switch to occur and archiving to happen while logging is also happening.

14) What is a control file used for?

Ans: The control file is used to keep information about the structure of the database. When the instance is started, the control file is used to identify the database files.

15) Can the instance be started without a control file?

Ans: Without the control file, DBMS will not know which datafiles to open. Oracle cannot start without a control file.

16) What can you do to help re-create the control file?

Ans: The ALTER DATABASE database BACKUP CONTROLFILE command is used to back up the control file. Using the TO TRACE qualifier generates the SQL statements necessary to reproduce the control file.

18) What is a rollback segment used for?

Ans: Just as the name indicates, a rollback segment is used to store change information that can be used in the event of a rollback. It is also used for read consistency.

19) How big is a rollback segment?

Ans: The rollback segment dynamically allocates and deallocates space. If the OPTIMAL value is set, the rollback segment will try to stay that size.

20) Should more than one user share a user account?

Ans: No, the user accounts are individual. Sometimes the DBA might be auditing the system and will be able to tell what users have been active and when. If many people use the same account, the DBA cannot tell who is using the system.

21) What are the server processes?

Ans: The server processes are the background processes or threads that perform tasks on behalf of DBMS.

22) What is a dedicated server process?

Ans: A dedicated server process has a one-to-one correlation between the user process and the server process. Each user process gets one server process.

23) What is a shared server process?

Ans: The shared server process handles more than one user process. The dispatcher queues the job and the shared server process executes it.

24) What are tables used for?

Ans: A table is the DBMS object that actually holds the data that is entered into the database. A table consists of rows of data that are defined by table columns.

25) What is a partitioned table?

Ans: A partitioned table uses the new Oracle feature, range partitioning. Data is stored in a location based on a range of data you have defined. This range can be numeric or a data format.

26) Why would you use a view?

Ans: Views can be useful to hide certain information from the end user. They can be used for security purposes or to simplify access to a distributed table or complex join.

27) What makes up an object?

Ans: An object consists of an attribute and a method.

28) What is a nested table?

Ans: A nested table is a table that appears as a column in another table.

29) What is a partitioned table?

Ans: A partitioned table is a table where the data is divided into smaller pieces based on the data itself.

30) How are tables partitioned?

Ans: DBMS supports range partitioning. The data is divided into various chunks based on ranges of data in one or more columns.

31) What is a view?

Ans: A view is a logical representation of a subset or superset of information in schema objects.

32) What is a synonym?

Ans: A synonym is simply an alias to another object in the database. This object can be a table, view, sequence, or program unit.

33) How are indexes important to an RDBMS?

Ans: An index provides a mechanism by which to find data quickly without having to read all records in a table.

34) What are some important factors involved in creating an index?

Ans: Several factors should be kept in mind when creating an index. First, an index is only useful when it is accessed. If you do not access the table using the columns that you have indexed in the WHERE clause, the index will not be used. Second, if the table is used primarily for historical data (many insertions, few selections), it is not a good candidate for indexing.

35) What makes a column a good candidate to be used in an index?

Ans: Columns that have many unique values are good candidates for indexing. If a combination of columns is unique, a complex index might be better.

36) What is a B*-tree index?

Ans: B*-tree index is a binary tree structure that can be used to find data quickly. A binary comparison is made and the tree is traversed based on that selection.

37) Can indexes be partitioned?

Ans: Yes. In fact, there are two ways to partition an index. A local partitioned index is an individual index that resides on a partition and indexes the data on that partition, whereas a global partitioned index is a single index over all of the partitioned data. A global partitioned index can in turn be partitioned itself.

38) What is an index-only table?

Ans: An index-only table is an index where the column data resides in the leaf block of the index. This schema can be very useful under certain conditions.

39) What is a bitmap index?

Ans: A bitmap index is an index where the index values are kept in a bitmap. A 1 indicates that the key value is present; a 0 indicates that the key value is not present.

40) How big can a sequence be?

Ans: A sequence can be as many as 38 digits in length--bigger than you will ever need.

41) Does an index need to be based on unique values?

Ans: An index does not need to be unique, but it can be.

42) Can an index be used to enforce uniqueness?

Ans: An index can be used to enforce uniqueness on a column or set of columns.

43) What is a standby database?

Ans: The standby database is a database that is constantly in recovery mode, recovering archived log files from the primary database. In the event of a failure, the standby database can immediately substitute for the primary database.

44) What is a read-only table snapshot?

Ans: A read-only table snapshot is a picture of a database or table that is copied to another system. The system from which the copy is made is called the master. The replicated system has the table or database available only in read-only mode.

45) What is an updateable snapshot?

Ans: An updateable snapshot is a replication method where both the master and the replication systems are updateable. The replicated systems synchronize themselves on a regular basis.

46) What is real time data replication?

Ans: Real time data replication causes all systems to be updated immediately after the change is made. This replication method needs all systems to be up and running for it to work effectively.

47) What is a master site?

Ans: The master site is the system from which the data is replicated. Because it is updateable and holds the original data, it is considered the master.

48) What is a snapshot site?

Ans: The snapshot site is the system that receives the snapshots. If the snapshot site is read-only, no updates are allowed and the site is considered a slave site.

49) What is the difference between a complete refresh and a fast refresh?

Ans: With a complete refresh, the entire snapshot is updated. With a fast refresh, only the changes are updated.

50) What is a standby database used for?

Ans: A standby database is a database that is constantly in recovery mode, recovering archive log files from the primary database. In the event of a failure, the standby database can immediately substitute for the primary database. The standby database is used to provide immediate restoration of service in the event of a primary system failure. In a matter of minutes or even seconds, this system can be operational.

51) Why would you use a read-only tablespace?

Ans: A read-only tablespace can be used to prevent users from updating critical data as well as to reduce recovery time, because no instance recovery is needed on a read-only tablespace.