Midterm

Which type of databases are not optimized for a single application?

NoSQL Databases

Data Warehouses

Embedded Databases

Enterprise Databases

Characteristics of data warehouses include all of the following EXCEPT:

Write Once / Read Mostly

Use of "Star" and "Snowflake" schemas

Batch Updates

Business Transactional Support

We will talk about "traversal" most commonly with regards to what type of database operation?

Transferrings data from a client to a server

Connecting two relationed instances of data

Connecting two federated databases

Searching a tree-based index structure from root to

Physical databases are common created during which part of the "waterfall" development process?

Conceptual modeling focuses on:

Creating data structures based on database models

Designing data architecture of applications

Identifying data objects and basic relationships between entities and values

Collecting external artifacts from stakeholders

Logical modeling focuses on:

Using patterns to design database application code

Designing queries for application transactions

Identifying data objects and basic relationships between entities and values

Creating data structures based on database models

Usually, databases are constructed before the detailed design and coding of database application because:

Allow for earlier migration of existing data

Allows parallel effort within team

Implications of external coupling

Easier to build databases than code

The relational database model has which of the following characteristics:

"Master" and "detail" optimized structures

Stored relationships that optimize traversals

Multiple "master" or top-level entities / classes

Ad-hoc traversals based on arbitrary relationships

Which of the following most likely be considered "databases"?

The characteristics of the ideal unique identifier include all of the following except:

User meaningful

Never changes once created

Univeral uniqueness

String data type

Which statement best describes "smart keys":

Smart keys use cryptographic hashes to uniqueness.

Smart keys are artificial keys encode natural data and serial numbers.

Smart keys are easy to use and remember by people.

Smart keys are synonym for compound keys.

Modern DBMS commonly support triggers and/or allow the cascading of changes made to unique identifiers over time.

True

False

The key difference between conceptual data modeling and logical data modeling is:

Logical data modeling focuses on structures compliant with a given database model

No difference - conceptual data models are logical data models

Logical data modeling apply rules of Normal Forms

The use of different notation tools (diagrams)

What holds (implements) the relationship that exists between instances of related entities in the a relational logical data model?

For relational logical data models, which of the following statement best describes the use of foreign keys:

Foreign keys will always be found in the "many" side entity in a "one-to-many" relationship between two entities

Foreign keys will always be found in the "one" side entity in "one-to-many" relationship between two entities.

Foreign keys will always be found in both entities associated with a "one-to-many" relationship between two entities.

Foreign keys are optional and chosen by the person creating the data model.

The cardinality of inheritance relationships is:

Always one-to-one between supertype and subtype

Always one-to-many between supertype and subtypes

Always many-to-many between supertype and subtype

Varies depending on the semantics of the data

A "one-to-many" relationship between two entities (such as X and Y, respectively) means:

Each instance of X may have one or more related instances Y, and each instance of Y has one and only instance X related to it.

Each instance of X must have at least one instance Y related to it, and each instance of Y must have at one and only one instance X related to it.

Each instance of X may have one or more related Y instances, and each Y must have at least one instance of X related to it.

Each instance of Y may have one or more related Y instances, and each instance Y may have at least one instance of X related to it.

You can choose assess a data model conforms to 3rd Normal Form only-- in other words, without consideration of whether or not it complies with other Normal Forms).

True

False

A logical data model that has been changed to break one of the rules associated with normalization is called:

A denormalized model

Trick question - not allowed to break Normal Form rules in a logical data model

An unnormalized model

A normalized model

Commonly, logical data models are modified until they comply with which normal form:

Which Normal Form is associated with evaluating partial key dependencies:

Which Normal Form is associated with evaluating transitive key dependencies:

Which Normal Form is associated with eliminating repeating data groups:

Each predicate in an Object Role Model represents what with respect to logical data models and normalization:

One or more attributes depending the n-ary of the relationship

Foreign key declaration

Primary key declaration

Functional dependency

Which of the following statement(s) best describes why primary keys of entities are important in normalization?

Represent data that cannot be updated

All non-key attributes must be functionally dependent on the primary key

Represent the only data that can be duplicated

Used to help define foreign keys to hold relationships

Microsoft SQL Server uses a "slot array" as part of its row storage structure. The "slot array" is used for:

Storing ROW IDs of each row on the page

Storing byte-offsets to row data

Holding NULL / NOT NULL flags for column values

Storing byte-offsets to column data

Oracle internally stores numbers in which of the following formats:

Binary Floats (single-precision / double-precision)

Unsigned binary integers (8-bit, 16-bit, 32-bit)

Variable-length packed decimal reals with mantissa and exponent

Signed binary integers (8-bit, 16-bit, 32-bit)

Based on overhead, what is estimated maximum number of rows that can be stored on one Microsoft Sql Server 8K database page?

Which DBMS(s) incur no overhead for "trailing NULLs"? (Columns in table definition occur in a declared order such as ID, Last Name, First Name, Middle Name, Phone, Fax, Address. Lets assume that all columns except ID are declared as NULL. Then "trailing NULLs" might be a case where a row had NULL for Phone, Fax, and Address. If Fax or Address was not a NULL value, then there would not be trailing NULLs no matter what other columns had NULL values for a given row.)

Microsoft SQL Server

No answer text provided.

MySql

Oracle

For MySql databases (using InnoDB data engine), how many bytes are required per field pointer in a row header?

1 byte

4 bytes

2 bytes

1 or 2 bytes that depends on the overall length of the row.

Which DBMS reserve space for fixed-length columns regardless of whether of not they are NULL?

No answer text provided.

Oracle

MySql

Microsoft Sql Server

Midterm

More Quizzes