A request to a Web server can include response data from a previous
request. The term we use for this type of association is data correlation.
Another commonly used synonymous term is dynamic data.
Interactions with an application are typically interrelated with one another.
Specifically, a request made to a Web server frequently includes data that
was returned to the Web browser in the response to a previous request. For
example, consider the following interaction with a Web-based application:
- A payroll clerk enters the URL for an application, which sends a login
prompt. When the clerk successfully logs in, the Web server returns a session
ID to the clerk's web browser along with a page indicating that login has
succeeded.
- The clerk clicks a link on the returned page that requests the Web server
to open the employee database search page. Unbeknownst to the clerk, the Web
browser sends with the request the session ID. Based on the session ID, the
Web server knows that the request comes from someone who is already logged
on and so returns the employee database search form. The clerk now performs
one or more name search operations in order to locate a specific employee.
Among other information, the Web server returns a photograph and a unique
ID. (Depending on application design, the employee ID can be visible or hidden.)
- The clerk clicks a link that requests the Web server to return the current
employee's payroll record. With this request, the Web browser sends both the
session ID, so that the Web server knows that the request comes from someone
who is logged on, and the employee ID, so that the Web server can locate and
return the correct information.
In this example, request 2 is dependent on request 1 and request 3 is
dependent on requests 1 and 2.
If you recorded these interactions in a performance test, prior to running
the test with multiple users, you would likely vary the test data: substitute
the user name/password values, the employee name search values, or both, with
values contained in datapools. When you ran the test, each test user would
return a different employee payroll record, based on the contents of the datapool(s).
In a generated performance test, where data in a request is dependent on
data contained in the response to a previous request, the request data is
substituted from the response data on which it depends. The term for this
internal tagging of response and request data is data correlation (or,
sometimes, dynamic data). Data correlation is required for correct
multiuser test execution with varied data.
A reference is a value in a test (usually in a response)
that can be used by a subsequent value (usually in a request) in the test.
When the test generator detects that a request value should be substituted
from a previous value, it designates the earlier value as a reference and
correlates the subsequent request value with the reference. This is referred
to as automated data correlation. But you can manually correlate
any two values in a test or unlink existing correlations. (You can also modify
or disable automated data correlation in the preference settings: click Window
> Preferences, expand Test, and click Performance
Test Generator.
To help you work with correlated data, the test editor uses color coding
and provides navigational aids:
- When you click a page, the page Details include a Test data table
listing datapool candidates (black lettering), datapooled data (green lettering)
and correlated data (red lettering) in the page. (To enable display or correlated
data, right-click any table and select Show References.)
You can navigate directly to a page request containing correlated data by
double-clicking the appropriate table row. You can also datapool correlated
data from this table, by clicking the row and then the Datapool
Variable button under the table.

- When you expand a page, a yellow highlight indicates page requests that
contain datapooled data or candidates (green lettering) or correlated data
(red lettering).

- When you click a highlighted request, datapool candidates (request values
that the test generator determined you were likely to substitute with datapool
values) are highlighted in light green, datapooled values in dark green, and
correlated data in red. If, as shown in the figure, you right-click a correlated
value, you can Go To its reference:

- References are highlighted in dark blue:
