True-False Questions
1.
A component is a type of design class.
Answer: False Page Reference: 336 Difficulty: Moderate
2.
Two-tiered architectures usually combine data management and
applications responsibilities into the same network layer.
Answer: False Page Reference: 338 Difficulty: Moderate
3.
Client-server applications usually involve the use of middleware.
Answer: True Page Reference: 338 Difficulty: Moderate
4.
HTML and JavaScript are often used to represent business rules in a
three-tier architecture.
Answer: False Page Reference: 339 Difficulty: Moderate
5.
Tight-coupling is an advantage of three-tiered architectures.
Answer: False Page Reference: 339 Difficulty: Moderate
6.
Architectures with more layers can support more scalability.
Answer: True Page Reference: 339-340 Difficulty: Moderate
7.
A single analysis class almost invariably maps to multiple design
classes.
Answer: False Page Reference: 340-341 Difficulty: Moderate
8.
Components are usually created by in-house staff for each application.
Answer: False Page Reference: 341 Difficulty: Moderate
9.
A package is a collection of classes.
Answer: True Page Reference: 341 Difficulty: Moderate
10. Third-party software components are
usually well tested and relatively error free.
Answer: True Page Reference: 341 Difficulty: Moderate
11. The COM standard is developed and
maintained by OMG.
Answer: False Page Reference: 342 Difficulty: Moderate
12. CORBA is designed specifically to
support software written in Java.
Answer: False Page Reference: 343 Difficulty: Moderate
13. RMI is a middleware that enables
the objects and methods of another Java program running on remote computers.
Answer: True Page Reference: 344 Difficulty: Moderate
14. The standard protocol for network
transmissions on the internet is SMTP.
Answer: False Page Reference: 346 Difficulty: Moderate
15. XML is superior to HTML for data
interchange.
Answer: True Page Reference: 346 Difficulty: Moderate
16. XML documents are typically
transported through the Internet by XML Schemas.
Answer: False Page Reference: 346-347 Difficulty: Moderate
17. COM, ASP, and XML/SOAP are
technologies incorporated into Microsoft’s .NET framework.
Answer: True Page Reference: 347-348 Difficulty: Moderate
18. Active X controls evolved from
technologies related to dynamic link libraries.
Answer: True Page Reference: 348 Difficulty: Moderate
19. ASP .NET scripts execute via
client browsers.
Answer: False Page Reference: 349 Difficulty: Moderate
20. High cohesiveness is a desirable
characteristic to strive for when creating packages.
Answer: True Page Reference: 352 Difficulty: Moderate
21. Tight coupling enhances the
design quality of object-oriented software systems.
Answer: False Page Reference: 354 Difficulty: Moderate
Multiple-Choice Questions
22. Design classes and ____________
are the two types of design elements.
a.
frameworks
b.
components
c.
packages
d.
tiers
Answer: b Page
Reference: 336 Difficulty: Easy
23. Which term refers to the process
of designing applications that embrace change and integrate smoothly with
internal and external processes?
a.
Application Change Development
b.
Business Process Reengineering
c.
Enterprise
d.
Application Change Management
Answer: c Page
Reference: 337 Difficulty: Moderate
24. An advantage of two-tiered
architectures over three-tiered architectures is that two-tiered architectures:
a.
have thinner clients involving less need for upgrades.
b.
need no middleware in order to connect to databases.
c.
have less network traffic between client and database server.
d.
allow application changes to be tested in a few workstations.
Answer: d Page
Reference: 339 Difficulty: Hard
25. The middle tier of a three-tier
architecture is primarily responsible for implementing:
a.
business rules.
b.
presentation logic.
c.
data access.
d.
network control.
Answer: a Page
Reference: 339 Difficulty: Moderate
26. Access to the functionality of
software components is provided by well-defined:
a.
business rules.
b.
interfaces.
c.
public attributes.
d.
Web services.
Answer: b Page
Reference: 341 Difficulty: Moderate
27. ____________ are reusable objects
that are distributed in binary form and support plug-and-play.
a.
Virtual machines
b.
Frameworks
c.
Packages
d.
Components
Answer: d Page
Reference: 341 Difficulty: Moderate
28. The Component Object Model (COM)
is a software component standard developed by:
a.
OMG.
b.
Sun.
c.
Microsoft.
d.
ANSI.
Answer: c Page
Reference: 342-343 Difficulty: Moderate
29. The Object Management Group’s
standard for distributed and heterogeneous networks of objects is called:
a.
JDBC.
b.
ActiveX.
c.
EJB.
d.
CORBA.
Answer: d Page
Reference: 343 Difficulty: Moderate
30. Remote Method Invocation (RMI) is
a technology that allows:
a.
Java programs running on different machines to communicate with each
other.
b.
designers to create components using UML models that automatically
generate Java code.
c.
programs written in different programming languages to communicate with
each other via an interface definition language.
d.
Web services to be sent between different applications via XML scripts.
Answer: a Page
Reference: 344 Difficulty: Moderate
31. Applet programs execute on:
a.
Web servers.
b.
application servers.
c.
client computers.
d.
mainframe hosts.
Answer: c Page
Reference: 344 Difficulty: Moderate
32. JBDC is a technology for
connecting:
a.
a Java applet to a browser.
b.
a Java program to a database.
c.
a Java server page to a Web server.
d.
two Java programs together for network communications.
Answer:
b Page
Reference: 344 Difficulty: Moderate
33. Which of the following is a
language whose syntax is made up of tags similar to those interpreted by web
browsers?
a.
IDL
b.
XML
c.
JavaScript
d.
CLR
Answer: b Page
Reference: 346 Difficulty: Moderate
34. Which of the following is a
text-based transport protocol supporting B2B web services?
a.
HTTP
b.
SOAP
c.
COM
d.
ADO
Answer: b Page
Reference: 346-347 Difficulty: Moderate
35. Email exchanges over the Internet
are governed by the ____________ protocol?
a.
FTP
b.
HTTP
c.
ASP
d.
SMTP
Answer: b Page
Reference: 346-347 Difficulty: Moderate
36. XML and SOAP are primarily used
to develop:
a.
Web services
b.
client-side scripts
c.
binary components
d.
APIs.
Answer: a Page
Reference: 347 Difficulty: Moderate
37. The framework standard provided
by Sun that competes with Microsoft’s .NET is called:
a.
OMG.
b.
CORBA.
c.
J2EE.
d.
SOAP.
Answer: c Page
Reference: 347 Difficulty: Moderate
38. A _____________ is a named
description of a problem and solution that can be applied to new contexts.
a.
pattern
b.
signature
c.
servlet
d.
bean
Answer: a Page
Reference: 353 Difficulty: Moderate
39. When specifying responsibilities
for design classes of a system, you should aim for:
a.
high coupling and high cohesion.
b.
high coupling and low cohesion.
c.
low coupling and high cohesion.
d.
low coupling and low cohesion.
Answer: c Page
Reference: 354 Difficulty: Moderate
40. In a UML design class diagram, a
plus sign indicates:
a.
an input method argument.
b.
a constructor method.
c.
public attribute visibility.
d.
a mathematical operation.
Answer: c Page
Reference: 354 Difficulty: Moderate
41. Design classes extend analysis
classes by adding all the following details EXCEPT:
a.
attribute data types.
b.
method arguments.
c.
visibility specifications.
d.
package collections.
Answer: d Page
Reference: 353-357 Difficulty: Moderate
42. Get methods are used for
accessing data in ____________ attributes.
a.
public
b.
private
c.
derived
d.
composite
Answer: b Page
Reference: 354 Difficulty: Moderate
Essay and Problem-Solving
Questions
43. Discuss the topic of layered architectures.
How does this relate to object-orientation? Describe single-tier, two-tier, and
three-tier architectures. Indicate how processing responsibilities are
distributed in each architecture, and compare the advantages and disadvantages
of each.
Answer:
In a layered architecture, a system is partitioned such that each
layer performs a specific type of functionality and communicates with its
adjoining layers. In this sense, the layers are like components with private
implementations and providing public interfaces. The single-tier architecture
is typified by a mainframe computer controlling presentation, business rules,
and database access. This architecture is very inflexible and rare for new
systems. Typically, only legacy mainframe-oriented systems and stand-alone PC
programs use this architecture. The
two-tier architecture is much more common and is often called a client-server
architecture. Typically, client-server architectures involve applications that
run on the client computers, while the server computer controls the database
and provides data access to the client applications. The interface is provided
by middleware such as ODBC. Since both the application and presentation logic
is performed on the client, this is called a thick client. Client-server
architectures are sufficient for simple applications, but have limitations.
Because the application logic runs on the client, this causes heavier demand on
client computers and requires application changes to be distributed to a large
number of client workstations. The three-tier architecture assigns only
presentation logic to the client, often via the browser interpreting HTML or
Javascript, leaving all business rule application logic to the middle tier of
application servers, which are in turn connected to the data layer on a
database server. This provides a loosely coupled system in which each layer’s
internal changes don’t affect other layers as long as the interfaces remain the
same. Because processing load is distributed across more layers, this
architecture is more scalable than single-tier or two-tier architectures.
Application changes only affect the middle layer computer, and don’t need to be
distributed to many clients as in the client-server model. However, three-tier
systems are more complex, require more interaction between different
proprietary layers, and involve more sophisticated network demands.
Page
Reference: 337-340 Difficulty: Moderate
44. What is middleware? How can use
of middleware ease construction of database-driven applications? Name at least
three middleware standards for database processing.
Answer:
Middleware
is software that provides one set of interfaces for connection to a client and
another set for connection to a server, thereby providing possibility of connections
between one of several clients and one of several servers. Very frequently
(although not always) application programs make use of one of the major
middleware packages for their database processing needs. Microsoft has
developed (and evolved) several database middleware technologies, including
Open Database Connectivity (ODBC), OLE DB, and ActiveX Data Objects (ADO). For Java-based applications, Sun provides
JDBC.
Page
Reference: 338, 344-346,
348-349 Difficulty: Moderate
45. Discuss the process of mapping
analysis classes to design elements. What types of transformations are there?
What additional information is added to analysis class diagrams as they evolve
into design class diagrams?
Answer:
An analysis class may map to a single design class, or be split
into multiple design classes. Or, many analysis classes can map into a single
design class, component, or package. Most entity classes map directly to design
classes, whereas control classes often split into multiple design classes for
modularity. Design classes are typically created in-house or available via a
third party class library. Boundary classes are often mapped to, or use,
components. Components are almost invariably third party products provided in
binary form. Design classes inherit the attribute and method signatures from
analysis classes, perhaps with some modifications. Attribute and method
definitions are refined with data types and visibility constraints. Method
definitions also include arguments.
Page
Reference: 340-341, 353-357 Difficulty: Moderate
46. What is a software component?
Describe the roles components play in systems, their means of communication
with other design elements, and other characteristics of components. What is
the relationship between a component and a class? Between a component and an
application? How are components similar
to and different from packages?
Answer:
A software component is a software element that with a
well-defined interface and functionality used for a specific purpose in a
plug-and-play manner by many applications. A component communicates to the
outside world through the messages defined in its interface, often called an
application programming interface (API). A single component may be made up of
several classes and interfaces.
Components are generally provided by third-party vendors as binary code
elements, using a variety of standards such as COM, ActiveX, Enterprise Java
Beans, and CORBA. A single component may be used by a wide number of
application programs. Components are similar to packages in that they both are
collections of classes and interfaces. However, whereas components are used in
a plug-and-play manner to provide functionality, packages are simply groupings
of classes according to functionality, acting more as a library repository than
a component of a system. Unlike components, packages do not, typically, have a
single well-defined function and interface.
Page
Reference: 341-342, 352-353 Difficulty: Moderate
47. What is a transaction? What are the
processing requirements in order to assure the integrity of a transaction?
Which technologies and services are available to facilitate transaction
processing in highly distributed, object-oriented systems?
Answer:
A transaction is an all-or-nothing unit of business work. This
means that all of the computer operations that make up a transaction must be
completed successfully. Thus, systems that support transactions should
guarantee four properties: atomicity, consistency, isolation, and durability (ACID).
Many products and standards include features supporting ACID. These include
Microsoft’s .NET framework and COM model, based on Microsoft Transaction
Services. Java’s EJB architecture, OMG’s CORBA, and XML/SOAP standards all
provide transaction management for distributed systems.
Page
Reference: 343, other pages Difficulty: Moderate
48. Identify, describe and compare
three leading binary-oriented component standards. For each standard cited,
identify the organization responsible for the standard, and discuss some of the
underlying technology and design elements. What are some of the limitations of
these standards?
Answer:
The three leading binary-oriented component standards are
Microsoft’s Component Object Model (COM), the Object Management Group’s (OMG)
Common Object Request Broker Architecture (CORBA), and Sun’s Enterprise Java
Beans (EJB). COM is the basic architecture that underlies most of Microsoft’s
.NET framework, and has extended into COM+, which includes a run-time
environment for middle-tier components and support for many Microsoft
Transaction Server functions. In addition, DCOM is distributed COM, providing
features for remote procedure calls in distributed systems. Components built
using the COM architecture run effectively in Microsoft’s IIS Web Server and
are frequently used in ASP scripts. COM objects do not, however, work well in
non-Microsoft platforms.
The CORBA standard was published by the Object Management Group, a
non-profit consortium that produces and maintains specifications for
interoperable enterprise applications. It was designed to allow distributed
objects, operating on heterogeneous networks, to interact with each other
seamlessly. Communication is managed by an Object Request Broker (ORB), which
directs requests from CORBA clients to CORBA servers. Thus method invocation by
the client does not require knowledge of the location of services. Applications
request the service, and the ORB finds the server that provides the service.
CORBA’s advantage is that heterogeneous platforms can interact with each other.
A site with legacy COBOL applications on a mainframe can interact with a
client-server Linux system, using ORBs and a layer written in Interface
Definition Language (IDL).
Sun’s EJB is written in Java and executes on any machine running a
Java Virtual Machine (JVM). It is a server-side component model addressing
issues involving management of distributed objects in a multi-tier
architecture. The EJB standard imposes naming conventions and other design
rules to construction of Java classes. Any Java class can be an Enterprise Java
Bean if it is designed according to these rules. EJB components can be invoked
using Java’s Remove Method Invocation (RMI).
Although each of these
standards has its strengths, all suffer from common limitations. They are
proprietary and binary, tightly coupled, and lock their data transfer on
underlying network and object infrastructures. None support the TCP/IP
protocol, but instead all rely on other interfaced layers (often via XML and
SOAP) to work in TCP/IP environments.
Page
Reference: 342-346 Difficulty: Moderate
49. How do XML and SOAP assist with
business-to-business commerce? What is XML, and what role does it play? What
are the similarities and differences between XML and HTML? What is SOAP, and
what role does it play? Finally, what are Web services, and how are they
related to XML and SOAP?
Answer:
Extensible Markup Language (XML) is a method for representing
structured data in a text file to facilitate data communication between
computer applications via the Internet. XML’s syntax, composed of tags and
attributes, is similar to HTML. But whereas HTML is limited with presentation
functionality, XML is primarily a structured data representation. Data is
defined in XML Schemas or Document Type Definitions. Simple Object Access
Protocol (SOAP) is a protocol for text-based representation of message passing
between systems. SOAP messages include an envelope, header, and body section.
The body section typically contains XML formatted data pertaining to the
message. Web services, a process and set of protocols for discovering software
as services across the Web. Specifically, Web services can use a data protocol like XML and a transport
protocol like SOAP as a foundation for
providing a standardized loose coupling for B2B transactions.
Page
Reference: 346-347 Difficulty: Moderate
50. What is a framework? Identify and
discuss two major application development frameworks. Give names and
descriptions of the technologies, programming languages, and component
protocols that underlie each of these frameworks. For each framework, identify
the technologies that support program compilation and execution, database
connectivity, and dynamic Web page creation.
Answer:
A framework is a collection of components, including code and
interfaces written in a specific language, that solves or helps build
applications. It provides the underlying services required for developing and
executing applications. Two major frameworks in the IT industry are Microsoft’s
.NET framework and Sun’s Java2 Enterprise Edition (J2EE). The .NET framework
integrates components built using the extended Component Object Model (COM+)
protocol, utilizing Active Server Pages (ASP .NET) for dynamic Web page construction,
and ActiveX Data Objects (ADO
Page
Reference: 347-349 Difficulty: Moderate
51. What is a responsibility? How are
responsibilities represented in sequence and class diagrams? Discuss three
guidelines that can help ensure proper allocation of responsibilities when
mapping analysis classes to design classes.
Answer:
Responsibilities form the behaviors assigned to an analysis class
based on its interface to outside classes. A responsibility is represented by a
message in a sequence diagram, and is shown as a method in an analysis class.
Responsibilities should be assigned to classes with three guidelines in mind.
First, behaviors should be performed by the information expert, that is, by the
class containing the information required for the behavior. Second, creation of
an object should be the responsibility of the classes that aggregate, contain,
or closely use the class of that object. Third, distribution of responsibility
should result in loose coupling and strong cohesion.
Page
Reference: 353-354 Difficulty: Moderate
52. In terms of design class
creation, what is visibility? What are
three different types of visibility? Which rules of thumb should be used when
assigning visibility to ensure object encapsulation? How can access to
attribute values be controlled?
Answer:
Visibility refers to the characteristics of an attribute or
operation that reveal whether it can be accessed directly. There are three
types of visibility, each of which can be applied to any attribute or operation
of a class. Public visibility means that any object external to the class can
access the attribute’s values or invoke the operation. With private visibility,
only other operations within the class can access the attribute or operation.
With protected visibility, operations from the class or its descendent classes
can access the attribute or operation. The principle of encapsulation implies
that attributes should be held private, and that controlled access to private
attribute values can be provided using public get- and set- methods.
Page
Reference: 354-355 Difficulty: Moderate
No comments:
Post a Comment