Deck 13: Distributed Databases

Question 1

Replication should be used when:

A) data currency and timing don't matter.
B) there is a highly heterogeneous network.
C) there are no or few triggers.
D) spreadsheet pages need to be linked.

Answer 1

C

Question 2

Snapshot replication is most appropriate for:

A) a data warehouse application.
B) an application where the processing of data follows a workflow across the business units.
C) a transaction processing system.
D) a spreadsheet application.

Answer 2

A

Question 3

________ is a design goal for a distributed database, which says a user does not need to know the location of data to use the data.

A) Location transparency
B) Location autonomy
C) Location dependency
D) Distributed locatability

Answer 3

A

Question 4

Which of the following is true about horizontal partitioning?

A) Data can be stored to optimize local access.
B) Data are less secure.
C) There is consistent access speed.
D) It only works with Macintosh systems.

Question 5

All of the following are disadvantages of data replication EXCEPT:

A) storage requirements.
B) reduced network traffic at prime time.
C) cost of updating.
D) complexity.

Question 6

Which of the following are business conditions that encourage the use of distributed databases?

A) Companies with less than 10 employees
B) Lack of Data sharing needs
C) Data communication reliability
D) Companies that only store data on spreadsheets

Question 7

________ is a design goal for a distributed database that says a site can independently administer and operate its database.

A) Remote autonomy
B) Local authority
C) Local autonomy
D) Location transparency

Question 8

Which of the following supports a simple path to other databases, without the benefits of one logical database?

A) Linked; differential
B) Gateways
C) Autonomous; cross-linked
D) Federated; nonautonomous

Question 9

Which of the following characterizes homogeneous environments?

A) Contains simple pathways
B) Same DBMS used at all locations
C) Some users require only local access
D) Cross-linked systems easily communicate

Question 10

Which of the following is true of data replication?

A) Reduced storage requirements
B) Slower response
C) Node decoupling
D) Requires Oracle

Question 11

Data replication allowing for each transition to proceed without coordination is called:

A) reliability.
B) fast response.
C) node decoupling.
D) node coupling.

Question 12

Which of the following environments uses the same DBMS at each node with a central or master DBMS coordinating database access across nodes?

A) Centralized; maximum
B) Centralized; minimum
C) Homogeneous; nonautonomous
D) Federated; nonautonomous

Question 13

A form of distributed database in which all data across a network are kept continuously updated, so a user can access any data anywhere on the network and get the same answer is called a(n) ________ distributed database.

A) synchronous
B) asynchronous
C) private
D) cloud-based

Question 14

Which of the following is true of distributed databases?

A) Less reliable
B) Better local control
C) Slower response time
D) Centralized

Question 15

One way to generate, store, and forward messages for completed transactions to be broadcast across a network is through the use of:

A) stored procedures.
B) triggers.
C) functions.
D) SQL statements.

Question 16

________ is an application that can effectively employ snapshot replication in a distributed environment.

A) Data cloning
B) Data replication
C) Distributed data
D) Data warehousing

Question 17

Simple paths to other databases without the benefits of one logical database are called:

A) freeways.
B) parkways.
C) gateways.
D) highways.

Question 18

Which of the following environments uses a different DBMS at each node and supports local databases for unique data requests?

A) Gateways; local
B) Centralized; minimum
C) Homogeneous; cross-linked
D) Heterogeneous; federated

Question 19

With a pull strategy of replication, the ________ node determines when a database is updated.

A) user
B) source
C) target
D) banner

Question 20

A(n) ________ is a database stored on multiple computers in multiple locations that are NOT connected by a data communications link.

A) distributed database
B) decentralized database
C) unlinked database
D) data repository

Question 21

A distributed database is NOT just a collection of files.

Question 22

All of the following are advantages of vertical partitioning EXCEPT:

A) efficiency.
B) security.
C) ease of querying.
D) easier to set up than horizontal partitioning.

Question 23

Distributed databases do not easily satisfy both transaction and analytical processing systems.

Question 24

A joining operation in which only the joining attribute from one site is transmitted to the other site is called a(n):

A) attribute linkup.
B) key data element transfer.
C) key data element join.
D) semijoin.

Question 25

A distributed database must:

A) keep track of data locations.
B) present a single logical database stays physically in one location.
C) require Oracle.
D) require SQL Server.

Question 26

The step in which a distributed database decides the order in which to execute the distributed query is called:

A) decomposition.
B) localization.
C) step optimization.
D) global optimization.

Question 27

An optimization strategy that allows sites that can update to proceed and other sites to catch up is called:

A) read-only commit.
B) lazy commit.
C) sequenced commit.
D) linear set up.

Question 28

A synchronized replication strategy has a(n) ________ reliability.

A) excellent
B) good
C) fair
D) medium

Question 29

________ ensures that a transaction is successfully completed or else it is aborted.

A) Failure protocol
B) TCP/IP protocol
C) Commit protocol
D) Go/No Go protocol

Question 30

A design goal for distributed databases that states that although a distributed database runs many transactions, it appears that a given transaction is the only one in the system is called:

A) linear run.
B) concurrency transparency.
C) transaction CPU transparency.
D) location transparency.

Question 31

A design goal for distributed databases to allow programmers to treat a data item replicated at several sites as though it were at one site is called:

A) location transparency.
B) replication accessibility.
C) replication transparency.
D) data accessibility.

Question 32

In a distributed database, a transaction that requires reference to data at one or more nonlocal sites is called a ________ transaction.

A) link
B) tight
C) global
D) loose

Question 33

A ________ allows a single SQL statement to refer to tables in more than one remote DBMS.

A) distributed UOW
B) join SQL statement
C) distributed request
D) data transfer command

Question 34

The sequence of instructions required to process a transaction is called the:

A) unit of work.
B) logical program unit.
C) rollback.
D) commit.

Question 35

With ________, all of the actions of a transaction are either committed or not committed.

A) location transparency
B) replication transparency
C) failure transparency
D) commit transparency

Question 36

Distributed databases make data sharing more difficult because of the data communication problems.

Question 37

With ________, users can act as if all the data were located at a single node.

A) location transparency
B) local autonomy
C) client-based control
D) location consistency

Question 38

Which of the following factors in deciding on database distribution strategies is related to autonomy of organizational units?

A) Organizational forces
B) Frequency of data access
C) Reliability needs
D) Organizational development

Question 39

A centralized strategy has ________ expandability.

A) excellent
B) good
C) poor
D) medium

Question 40

An integrated partition strategy is ________ to manage.

A) excellent
B) difficult
C) easy
D) medium

Question 41

The overriding principle in distributed database design is that data should be stored at one central site.

Question 42

Horizontal partitioning offers increased efficiency because data are close to use areas.

Question 43

One way to generate messages for near-real-time replication is through the use of database triggers.

Question 44

With an asynchronous distributed database, all data across the network is continuously kept up to date.

Question 45

A distributed database is a single logical database spread physically across computers in multiple locations that are connected by a data communications link.

Question 46

Local autonomy means that data are not accessible from remote sites.

Question 47

Replicated databases usually result in tightly coupled nodes.

Question 48

Location efficiency enables a user to access data without knowing exactly where the data are located.

Question 49

Applications such as decision support or data warehousing often do not require current data and are supported by periodic snapshots.

Question 50

In a homogeneous environment, the same DBMS is used at each location.

Question 51

A distributed database may require more costly and complex software than a centralized database.

Question 52

Replication should NOT be used if timely updates are important.

Question 53

Combining data across vertical partitions is generally more difficult than combining across horizontal partitions.

Question 54

Users of distributed database systems always need to know the location of the data.

Question 55

Applications that can tolerate out-of-date data are not the best candidates for data replication.

Question 56

Shared ownership of data is most appropriate as business activities move across time zones.

Question 57

In a heterogeneous environment, not all users use the same DBMS.

Question 58

When separate copies of data are stored at each of two or more sites, it is called replication.

Question 59

Replication is often used for noncollaborative data.

Question 60

An environment in which data are distributed across all nodes, and there is a global schema, is called a(n) heterogenous environment.

Question 61

With a two-phase commit strategy for synchronizing distributed data, committing a transaction is faster than if the originating location were able to work alone.

Question 62

Two-phase commit is an algorithm for coordinating deletions in a distributed database management system.

Question 63

Organizational forces such as funding availability drive the choice of distributed database strategy.

Question 64

A transaction that requires only reference to data that are stored at the site is a remote transaction.

Question 65

In a push replication strategy, the target, not the source, controls when a local database is updated.

Question 66

A local transaction is one that requires reference only to data that are stored at the site where the transaction originates.

Question 67

A distributed DBMS must provide consistency among copies of data across the remote sites.

Question 68

Failure transparency is a design goal for distributed databases that guarantees that all the actions associated with a transaction are either committed or none of them are committed.

Question 69

Concurrency transparency allows each transaction to appear as if it were the only activity in a distributed database management system.

Question 70

A transaction that requires reference to data at one or more nonlocal sites to satisfy the request is a global transaction.

Question 71

The reliability of a decentralized distributed database system with independent partitions is poor.

Question 72

The purpose of time-stamping is to allow transactions to be processed in parallel.

Question 73

A replicated database with snapshots is very manageable.

Question 74

In general, data should be located away from the applications that use those data.

Question 75

Horizontal partitioning is implemented by placing some rows of a table at one site and other rows at another site.

Question 76

Replication transparency allows a programmer or user to treat a replicated data item as if it were a single item at a single node.

Question 77

Location translucency enables users and programmers to treat data as if it were located at one node even though it may move from node to node.

Question 78

Location transparency is used to inform users of where data are located within the distributed database.

Question 79

Scalability is the ability to grow, reduce in size, and become more heterogeneous as the needs of the business change.

Question 80

A transaction manager is an algorithm that coordinates updates or aborts them.