Database Management Systems

A conformed dimension is a dimension table with the same meaning and content across multiple fact tables in a data warehouse. It enables drill-across analysis by allowing different fact tables to be joined through the shared dimension, ensuring consistency in reporting. Option A is incorrect as conformed dimensions are not created together but are designed to be compatible. Option B is wrong as all dimensions have primary keys. Option C is vague.

A snowflake schema normalizes dimension tables into multiple related tables to reduce redundancy and save storage space. Unlike star schemas where dimensions are denormalized, snowflake schemas break down dimensions into hierarchical levels (e.g., product → product category → product type). Option A incorrectly suggests facts are normalized (they're not - facts remain denormalized). Option B is wrong because fact and dimension tables are not joined together in 3NF - only dimensions are normalized.

Dimensions contain descriptive attributes that provide context for facts in the data warehouse. They are subject-oriented (organized around business entities like customer, product, time) and non-volatile (data doesn't change once loaded). Options A and D are wrong because dimensions don't contain facts. Option B is incorrect as dimensions contain their own descriptive data, not primarily data about other dimensions.

Cardinality in Framework Manager defines the numerical relationship between two tables (1:1, 1:N, N:1, or N:N) regardless of table type. It can represent relationships between fact and dimension tables (common in star schemas), between dimension tables (when dimensions relate to each other), or between any other tables in the model. Cardinality affects how Framework Manager generates SQL and handles joins. Option D is incorrect because cardinality is defined per relationship, not by data source.

Cognos Report Studio provides multiple data container types for displaying data: Lists show tabular data, Crosstabs display matrix/pivot tables, Charts create visual representations, and Conditional Blocks show content based on specified conditions. All four listed options are valid data containers in Cognos, making "All the above" the correct answer.

OLAP (Online Analytical Processing) structures are designed for fast analysis of multidimensional data from multiple perspectives. While OLAP cubes can aggregate data from various sources for near real-time analysis, option C best describes their core analytical capability. Options A and B describe traditional dimensional modeling, while option D describes table join configuration, not OLAP's defining characteristic.

The SQL shows a SELECT statement with two tables in the FROM clause but no JOIN condition or WHERE clause linking them together. This syntax produces a Cartesian product (cross join), where every row from the Manager table is combined with every row from the Product table. Cross joins are rarely intentional in reporting and typically indicate missing join conditions.

The LGWR (Log Writer) process writes redo log buffer to redo log files when: (A) the log buffer is one-third full, (B) a transaction is committed (ensuring durability), and (C) every three seconds (timeout-based write). Option D (redo log file switch) triggers LGWR to write to the new log file but is not one of the standard conditions for writing to the current log file. These mechanisms ensure that redo information is regularly persisted for database recovery.

The statement that C-EMS Oracle database is configured as NO ARCHIVELOG mode is false. ARCHIVELOG mode is essential for production databases as it enables point-in-time recovery and online backups. NO ARCHIVELOG mode would severely limit recovery options and is not appropriate for a production database environment. The correct configuration would be ARCHIVELOG mode.

A tablespace is the logical unit of storage in Oracle that combines one or more datafiles. Segments (A) are database objects like tables or indexes stored within a tablespace, blocks (B) are the smallest unit of data storage, and views (D) are virtual tables. Tablespaces provide a layer of abstraction between physical datafiles and logical database objects.

Rollback transaction data is stored in Undo data files (or Undo tablespace) in Oracle. This undo information is used to rollback transactions, provide read consistency, and enable Flashback features. System (A) tablespace contains data dictionary, Sysaux (B) contains auxiliary data, and Index (D) files store indexes - not transaction rollback data.

The block is the smallest logical unit in Oracle database storage. Multiple contiguous blocks form an extent, multiple extents form a segment, and segments are stored in tablespaces. Blocks are the fundamental unit of I/O operations.

The control file contains the database name, physical locations of all data files and redo log files, and critical metadata like checkpoint information. It's essential for mounting, opening, and recovering the database during startup.

The parameter file (init.ora or SPFILE) contains all initialization parameters like memory settings (SGA/PGA), process limits, maximum connections, and database configuration. These parameters are read during instance startup to configure the database behavior.

SMON (System Monitor) performs instance recovery automatically after a crash. It rolls forward committed changes from online redo logs and rolls back uncommitted transactions to bring the database to a consistent state. SMON also coalesces free space in tablespaces.