Topic: Te s t D a t a

Message

   Te s t D a t a

Test data
is the fuel for virtually all test cases. Because it is impossible to
test all data, careful consideration must be given to test-data
generation. Data should be analyzed and separated into classes based on
likelihood of occurrence and whether or not the data occurs on a
boundary condition. In general, the test data classes should be
representative of at least three basic classes of data—the maximum, the
minimum, and the most common, or average class. 

The easiest
way to visualize these three data classes is to map their occurrence to
a standard bell curve func-tion,  as shown in Figure 4. The large
shaded area represents the most commonly occurring examples. The small
shaded areas at the extremes of the curve represent the minimum and
maximum cases. By select-ing data that is represented in reality, by
this relationship, test data coverage is assured for almost 90 percent
of the area covered by the curve. 

Figure 4. Testing for the Maximum Data Class

The
minimum data class, as depicted by the bell curve, characterizes the
smallest data sets that can be manip-ulated by the product component
and exercises the minimum boundary condition. The maximum data class
characterizes the largest data set that can possibly be handled by the
product compo-nent and exercises the maximum boundary condition. The
most common data set represents the data that has the most likely
chance of occurrence. Using this approach, the product component can be
thoroughly tested without having to exercise every data sample. By
choosing test data samples from each test data class, broad coverage is
achieved without invest-ing unrealistic amounts of time.

Author	Message
tanushree Senior Member Joined: 04Apr2007 Online Status: Offline Posts: 2160	Topic: Te s t D a t a Posted: 17Oct2007 at 3:41am
	Te s t D a t a Test data is the fuel for virtually all test cases. Because it is impossible to test all data, careful consideration must be given to test-data generation. Data should be analyzed and separated into classes based on likelihood of occurrence and whether or not the data occurs on a boundary condition. In general, the test data classes should be representative of at least three basic classes of data—the maximum, the minimum, and the most common, or average class. The easiest way to visualize these three data classes is to map their occurrence to a standard bell curve func-tion, as shown in Figure 4. The large shaded area represents the most commonly occurring examples. The small shaded areas at the extremes of the curve represent the minimum and maximum cases. By select-ing data that is represented in reality, by this relationship, test data coverage is assured for almost 90 percent of the area covered by the curve. Figure 4. Testing for the Maximum Data Class The minimum data class, as depicted by the bell curve, characterizes the smallest data sets that can be manip-ulated by the product component and exercises the minimum boundary condition. The maximum data class characterizes the largest data set that can possibly be handled by the product compo-nent and exercises the maximum boundary condition. The most common data set represents the data that has the most likely chance of occurrence. Using this approach, the product component can be thoroughly tested without having to exercise every data sample. By choosing test data samples from each test data class, broad coverage is achieved without invest-ing unrealistic amounts of time. Post Resume: Click here to Upload your Resume & Apply for Jobs

	IP Logged

suresh5003 Newbie Joined: 18May2009 Online Status: Offline Posts: 1	Posted: 19May2009 at 12:23am
	< ="Content-" content="text/; charset=utf-8">< name="ProgId" content="Word.">< name="Generator" content="Microsoft Word 12">< name="Originator" content="Microsoft Word 12"><> Test Data Generation becomes the heart of the software testing, because it is difficult for the software testers to collect the data and used for testing purpose Recently I have seen a tool called TestersDesk.com which is used specially for Test Design and Test Data Generation Tools, using this tool all the organizations and software Testers are generating their data for testing purpose I can explain with good example Size-Based File Generator Size-Based file Generator is used to generate the Different file sizes for testing the download and uploading fields of different sizes It will be used in Naukri.com, Media players, etc In Naukri.com different people will upload different file size of Resume we don’t exactly what is the Maximum file size it will upload, for that we can’t create the different file sizes manually to check the maximum size of Naukri, to do this work with in seconds this tool is very beneficial Size-based File Generator generates windows batch files that can successively create files of the specified size and type. This is also used when we are testing the Media players like for checking the file size of songs
	Suresh
	IP Logged


One Stop Testing Forum : Software Testing @ OneStopTesting : Test Cases @ OneStopTesting