Chapter 12 Client/Server Database and Distributed Database
12-1
Last Update: November 15, 2011 — 7PM
Chapter 12: Client/Server Database and Distributed Database
Multiple Choice
1. All of the following are true about a two-tiered client/server LAN, except ____.
a. a gateway computer can link the LAN to other LANs or to the Internet
b. data can reside on a client PC’s hard drive or at the server
c. searching for data is location transparent
d. a printer on the LAN is available to all of the client PCs
e. data transparency can extend to a mainframe external to the LAN
2. In the database server approach to client/server database, when a client computer on
the LAN needs to query, update, or otherwise use a file on the server ____.
a. the entire file must be sent from the server to the client
b. the server can tend to become a bottleneck
c. the actual database search is performed at the client
d. the initial keyboard and screen interaction processing is performed at the server
e. the clients tend to require a great deal of processing power
3. Compared to the file server approach, the database server approach to client/server
database _____.
a. generates more network data traffic
b. requires more processing power at the server
c. requires a gateway connection to a mainframe computer
d. spreads security control through all of the client PCs
e. requires “fatter” client PCs with more processing power
4. In a three-tier approach to client/server database ____.
a. three of the client PCs on a LAN each constitute a tier
b. three LANs connected through gateway computers each constitute a tier
Chapter 12 Client/Server Database and Distributed Database
12-2
c. data location transparency runs through all three tiers
d. data can be stored at the tiers in the LAN but not at the one outside of it
e. the file server approach rather than the database server approach must be used
5. Consider a distributed relational database in which there is no data replication. All of
the following are true, except ____.
a. all of the table on the network is still available even if one site that has some
tables goes down
b. every table can be placed at the site at which it is most heavily used
c. if one site goes down then the data at that site cannot be accessed there or
anywhere else on the network
d. the integration of two tables that are not located at the same site will require a
distributed join
e. if one site goes down at least some of the database’s tables will still be accessible
6. If data replication is permitted in a distributed database network ____.
a. one site must contain a copy of every table
b. the maximum number of copies that can be made of any one table is equal to half
the number of computers (sites) on the network
c. distributed joins are never necessary
d. copies of tables can be located at the sites that most heavily use them
e. a copy of every table must be placed at every site
7. All of the following are principles of data replication in a distributed database
network, except ____.
a. place a copy of every table at a central node to avoid having a bottleneck
b. place copies of tables at the sites that use them most heavily in order to minimize
telecommunications costs
c. ensure that there are at least two copies of important or frequently used tables to
realize the gains in availability
d. limit the number of copies of any one table to control the security and concurrency
issues
e. avoid any one site becoming a bottleneck to maintain performance
Chapter 12 Client/Server Database and Distributed Database
12-3
8. If a distributed database network with data replication has one node that has a copy of
every table in the database ____.
a. the problem of data concurrency is eliminated
b. the issue of one site becoming a bottleneck is eliminated
c. the problem of data security is eliminated
d. at least one other copy of each table must also be stored elsewhere on the network
e. the problem of distributed joins is eliminated
9. If the nature of the distributed replicated data and of the applications that use it can
tolerate retrieved data not necessarily being up-to–the-minute accurate then a(n) ____
approach to updating replicated data can be used.
a. mirrored
b. duplicated
c. synchronous
d. asynchronous
e. versioning
10. The two-phase commit process can be used when the ____ approach to updating
distributed replicated data is required.
a. mirrored
b. duplicated
c. synchronous
d. asynchronous
e. versioning
11. The two phases of the two-phase commit process are ____ and commit.
a. copy
b. prepare
c. de-commit
d. replicate
Chapter 12 Client/Server Database and Distributed Database
e. concur
12. During the prepare phase of the two-phase commit____.
a. only the site issuing the update request writes the update to its database
b. all of the involved sites write the update to their logs
c. all of the involved sites write the update to their databases
d. all of the sites except the site issuing the update request write the update to their
logs
e. all of the sites except the site issuing the update request write the update to their
databases
13. If all went well during the prepare phase of the two-phase commit, then during the
commit phase ____.
a. only the site issuing the update request writes the update to its database
b. all of the involved sites write the update to their logs
c. all of the involved sites write the update to their databases
d. all of the sites except the site issuing the update request write the update to their
logs
e. all of the sites except the site issuing the update request write the update to their
databases
14. A distributed join is never necessary ____.
a. if distributed data is not replicated
b. if distributed data is replicated
c. if a copy of every table is kept at one site in a distributed database network
d. if there are at least two copies of every table in a distributed database network
e. if there are at least two copies of every table involved in the join in a distributed
database network
12-5
15. If node A on a distributed database network issues a query that requires tables at
nodes B, C, and D, the least likely site for the join to take place is node ____.
a. A
b. B
c. C
d. D
e. E
16. Horizontal partitioning is advisable on a distributed database network ____.
a. when particular records of a table are used at some network sites more than at
others
b. when all records of a table are used at some network sites more than at others
c. when different sites on the network require different columns of the table
d. when all of the columns of a table are used at some network sites more than at
others
e. None of the above.
17. In general, the directory of a distributed database ____.
a. should be stored at every site on the network because it tends to be updated
frequently
b. should be stored at every site on the network because it tends to be updated
infrequently
c. should be stored at only a limited number of sites on the network because it tends
to be updated frequently
d. should be stored at only a limited number of sites on the network because it tends
to be updated infrequently
e. should be stored at only one site on the network
18. All of the following are true of a centralized database on a network, except ____.
a. the site with the database can become a bottleneck
b. there is no need for a distributed directory since all of the data is in one place
c. the single site arrangement provides a high degree of security, concurrency, and
backup and recovery control
Chapter 12 Client/Server Database and Distributed Database
d. all data accesses from other than the site with the database incur
telecommunications costs
e. it uses a two-phase commit process
19. All of the following are true when dispersing tables on a distributed database network
without data replication or partitioning, except ____.
a. it may require distributed joins
b. it permits local autonomy over the data
c. it permits improved data availability because portions of the database are
available even if one or some of the sites are down
d. one site can tend to become a bottleneck
e. it requires a distributed directory
20. All of the following are true when there is targeted data replication on a distributed
database network, except ____.
a. there is no need for distributed joins because one site must have a copy of the
entire database
b. a two-phase commit is necessary if all of the replicated copies of a table must be
kept up-to-date all of the time
c. there is improved availability because of the data replication
d. communications costs for read-only data access are reduced because copies of
tables can be located at multiple sites that most heavily use them
e. it requires a distributed directory
True/False
1. What client/server database and distributed database have in common is that in both
cases the databases can be located other than “in the computer itself.”
2. The term “client” in a local area network refers to the larger computer with the shared
database that all of the PCs on the network can access.
3. In a two-tiered client/server arrangement, data can be found on each PC’s hard drive
and data can be in a database at the server.
4. In the database server approach to client/server database, when a client computer on
the LAN needs to query, update, or otherwise use a file on the server, the entire file
must be sent from the server to that client.
5. In the database server approach to client/server database, the actual database
processing is done at the server.
6. In the two-tier approach to client/server database, location transparency means that
the person issuing a query at a client does not have to know whether the data is at the
client or at the server.
7. One use of the term three-tier approach to client/server database is that data sought
from one of the client PCs can be on any one of a succession of three other PCs on
the LAN.
12-8
8. One use of the term three-tier approach to client/server database is that data sought
from one of the client PCs can be on the client’s hard drive, at the LAN server, or at a
mainframe beyond the LAN.
9. In one use of the term three-tier approach to client/server database the three tiers are
the client PCs, servers known as “application servers” and other servers known as
“database servers.”
10. When data is distributed in a distributed database arrangement, the database is no
longer considered to be a single, logical database.
11. A distributed database management system manages the distributed data as a single,
logical database.
12. In a network environment, a centralized database provides local autonomy over the
data.
13. Distributed data with no replication provides a high degree of availability.
14. Placing a complete copy of the database at every site on the distributed network
maximizes availability and minimizes concurrent update problems.
Chapter 12 Client/Server Database and Distributed Database
15. If in a distributed database network, a copy of the entire database is kept at the
headquarters site and each table is replicated exactly once at any one of the other sites,
the application environment can continue to function even if the headquarters site goes
down.
16. If in a distributed database network, a copy of the entire database is kept at the
headquarters site and each table is replicated exactly once at any one of the other sites,
there is no reason for the headquarters site to ever become a bottleneck.
17. One of the principles of data replication in a distributed database is to place copies of
tables at the sites that use them most heavily in order to minimize telecommunications
costs.
18. One of the principles of data replication in a distributed database is to ensure that there
are at least two copies of important or frequently used tables to realize the gains in
availability.
19. One of the principles of data replication in a distributed database is to maximize the
number of copies of any one table to control the security and concurrency issues.
20. If the nature of the data and of the applications that use it can tolerate retrieved data not
necessarily being up-to–the-minute accurate, then several “synchronous” approaches to
updating replicated data are appropriate.
21. The two-phase commit process is used for asynchronous updating of replicated data.
22. The two-phase commit process is used for replicated data that can tolerate having its
table copies not being up-to-date all of the time.
23. The commit phase is the first phase of the two-phase commit process.
24. In the prepare phase of the two-phase commit process, the updated data is written to a
log at each site but not to the database at each site.
25. In the commit phase of the two-phase commit process, either the process is aborted or
each site writes the update from their log to their database.
26. The more volatile the data in the database is, the more attractive is the two-phase
commit type of synchronous procedure for updating replicated tables in the distributed
database.
27. A distributed join is required when the tables needed for a join query do not all reside
at any one site on the distributed network.
28. In a distributed join situation, the only way that the join can be accomplished is to
transmit the records involved from their sites to the initiating site where the join must
take place.
29. The purpose of partitioning a table on a distributed network is to locate records of the
table at sites at which they are most frequently used.
30. Distributed directories are necessary in a distributed database environment in order to
achieve location transparency.
31. An advantage of a centralized database is that there is no need for a distributed
directory since all of the data is in one place.
32. Dispersing tables on a distributed network without replication avoids the need for a
distributed directory.
12-12
33. An advantage of targeted data replication in a distributed database network is reduced
communications costs for read-only data access.
34. Partitioned tables in a distributed database network tend to decrease local autonomy
over data.
Problems
1. Consider the following relational database for the Central Zoo. Central Zoo wants to
maintain information about its animals, the enclosures in which they live, and its
zookeepers and the services they perform for the animals. In addition, Central Zoo has a
program by which people can be sponsor of animals. Central Zoo wants to track its
sponsors, their dependents, and associated data.
Each animal has a unique animal number and each enclosure has a unique enclosure
number. An animal can live in only one enclosure. An enclosure can have several
animals in it or it can be currently empty. A zookeeper has a unique employee number.
Every animal has been cared for by at least one and generally many zookeepers; each
zookeeper has cared for at least one and generally many animals. Each time a
zookeeper performs a specific, significant service for an animal the service type, date,
and time are recorded. A zookeeper may perform a particular service on a particular
animal more than once on a given day.
A sponsor, who has a unique sponsor number and a unique social security number,
sponsors at least one and possibly several animals. An animal may have several
sponsors or none. For each animal that a particular sponsor sponsors, the zoo wants to
track the annual sponsorship contribution and renewal date. In addition, Central Zoo
wants to keep track of each sponsor’s dependents. A sponsor may have several
dependents or none. A dependent is associated with exactly one sponsor.
News flash!!! Central Zoo now has branches in four cities: Miami (also the Zoo’s
headquarters), Tucson, Detroit, and Cleveland. The database structure remains the same
except that the Location attribute of the ENCLOSURE relation now indicates in which
of the four cities the enclosure is located.
Chapter 12 Client/Server Database and Distributed Database
12-13
Central Zoo has decided to reconfigure its database as a distributed database among its
four locations. Distributed and replicated among these four locations, the tables have the
following characteristics:
ENCLOSURE consists of 4,000 records and is used most heavily in the cities in which
the individual enclosures are located.
ANIMAL consists of 50,000 records. The entire table is used in the Miami
headquarters, in addition to which the records concerning individual animals are used
frequently in the locations at which the animals are housed.
ZOOKEEPER consists of 1,500 records with a situation similar to the ANIMAL table.
CARES FOR consists of consists of 5,000 records with a situation similar to the
ANIMAL table.
SPONSOR consists of 10,000 records and is used mostly at the Miami headquarters.
CONTRIBUTION consists of 30,000 records with a situation similar to the SPONSOR
table.
DEPENDENT consists of 90,000 records with a situation similar to the SPONSOR
table.
Design a distributed relational database for Central Zoo. Justify your placement,
replication, and partitioning of the tables.
Enclosure
Number
Type
Location
Size
Date
Built
ENCLOSURE Relation
Animal
Number
Species
Animal
Name
Country
Of Birth
Weight
Enclosure
Number
ANIMAL Relation
Employee
Number
Employee
Name
Title
Year
Hired
ZOOKEEPER Relation
Animal
Number
Employee
Number
Service
Type
Date
Time
CARES FOR Relation
Sponsor
Number
Social Security
Number
Sponsor
Name
Address
Telephone
SPONSOR Relation
Animal
Number
Annual
Contribution
Renewal
Date
CONTRIBUTION Relation
Chapter 12 Client/Server Database and Distributed Database
12-14
Sponsor
Number
Dependent
Name
Relationship
Date of
Birth
DEPENDENT Relation
2. Consider the following relational database for Grand Travel Airlines.
Grand Travel Airlines has to keep track of its flight and airplane history. A flight is
uniquely identified by the combination of a flight number and a date. Every
passenger who has flown on Grand Travel has a unique passenger number. For a
particular passenger who has taken a particular flight, the company wants to keep
track of the fare that she paid for it and the date that she made the reservation for it.
Clearly, a passenger may have taken many flights (he must have taken at least one to
be in the database) and every flight has had many passengers on it.
A pilot is identified by a unique pilot (or employee) number. A flight on a particular
date has exactly one pilot. Each pilot has typically flown many flights but a pilot may
be new to the company, is in training, and has not flown any flights, yet. Each
airplane has a unique serial number. A flight on a particular date used one airplane.
Each airplane has flown on many flights and dates, but a new airplane may not have
been used at all, yet.
News flash!!! Grand Travel Airlines has decided to reconfigure its database as a
distributed database among its four “hub” airport locations: Boston (which is also its
headquarters city), Seattle, Nashville, and Columbus, Ohio.
Distributed and replicated among these four locations, the tables have the following
characteristics:
Basically, all of these tables are accessed by all four locations on a fairly equal basis,
except that processing at the Boston headquarters tends to be heavier than at the other
three hubs. Replication is used primarily for availability purposes in case one of the
Chapter 12 Client/Server Database and Distributed Database
12-15
sites crashes and the replicated copies are dispersed to avoid creating a bottleneck at any
one site.
Pilot
Number
Pilot
Name
Date of
Birth
Date of
Hire
PILOT Relation
Flight
Number
Date
Departure
Time
Arrival
Time
Pilot
Number
Airplane
Number
FLIGHT Relation
Passenger
Number
Passenger
Name
Telephone
Number
PASSENGER Relation
Flight
Number
Date
Passenger
Number
Fare
Reservation
Date
RESERVATION Relation
Airplane
Number
Passenger
Capacity
Year
Built
Manufacturer
AIRPLANE Relation
PILOT consists of 2,000 records and is located in Boston and Nashville.
FLIGHT consists of 1,500 records and is located in Columbus and Seattle.
PASSENGER consists of 250,000 records and is located in Columbus and Seattle.
RESERVATION consists of 1,500,000 and is located in Boston and Nashville.
AIRPLANE consists of 450 records and is located only at headquarters in Boston.
Assume that the telecommunications costs among the cities are all about the same.
Develop and justify a plan for solving the following queries.
a. A query is issued from Nashville to list the names and addresses of the passengers
who have reservations for a particular flight on a particular date one month from
now.
b. A query is issued from Nashville to get a list of the serial number, model, and
passenger capacity of the airplanes used on a particular flight (a flight that
operates daily) during the last two months.
c. A query is issued from Boston to get a list of the serial number, model, and
passenger capacity of the airplanes used on a particular flight (a flight that
operates daily) during the last two months.
Answer