Dynamo Scenario Server uses recorders to collect information for charts. Recorders are a type of scenario that is specifically designed to track data; the elements in a recorder tell the system to watch for certain site or visitor behavior, and then store information about that behavior in a given dataset.
The system comes with a set of predefined recorders that correspond to the functions of the different parts of the ATG Dynamo product suite and are used in the standard charts that come with system. You can reuse these recorders in custom charts by copying them and altering them as necessary. For more information, see
Creating Custom Charts in the dynamo documentation.
To view these recorders, select Scenarios > Scenarios, and then select the recorder you want to view from the Recorders folder at the top of the window.
The following shows the standard DAS recorder (stored as the file das.sdl):
The recorder contains two segments that work as follows:
A dataset is a named subset of an existing database. The dataset definition that appears in a Dynamo Scenario Server recorder contains the following information:
The name of the dataset (the dataset ID). The ID is a unique string that the system uses to identify the dataset in the database. When it is time to create a specific chart, the system uses the ID to find the data that it stored for that chart.
The sampling rate for the dataset. The sampling rate tells the system how much of the data that it tracks to store for use in the chart. This feature is useful for high traffic sites; it allows you to view meaningful reports without storing excessive amounts of data. For example, suppose you want to create a chart that identifies the most popular content items on a Web site. The site receives over ten thousand visits per day, and each visitor displays several content items. If you created a chart that stored every instance of an item that is displayed, you might store hundreds of thousands of pieces of data every day. You decide it is safe to assume that the most popular content items will be the same for 25% of your site visitors as they would be for 100%, so, rather than storing every instance of the View Content Item event, you change the sampling rate to store only 25% of the instances. You see the same results in the chart, you have stored less data, and the chart may also display more quickly on your screen.
You can change the sampling rate for a dataset through the Dataset window in the Dynamo Control Center. To display this window, select Analysis from the main Dynamo Control Center window, and then select Datasets.
The sampling rate options you can select are as follows:
All events that the system tracks
A percentage of tracked events
All tracked events for a percentage of sessions
A mapper. Mappers are .xml files that tell the system which specific aspects of the event to capture, and where to store the data. For example, in a recorder that tracks View Content Item events, the mapper tells the system exactly which information about the act of viewing content you are interested in - the name of each content item displayed, the name of the person who viewed the content, the age group of that person, and so on. (The range of information available to track depends on the attributes that exist in your content and profile repositories.)
The mapper tells the system which aspects of an event to capture by including a list of database rows and columns. This list tells the system which data to collect and where to store it.
The mapper in the DAS Startup recorder example shown above looks like this:
Each of the standard recorder scenarios that come with Dynamo Scenario Server contains a predefined dataset definition and mapper. Note also that different dataset definitions can share the same mapper.
As described above, you use charts to track specific information about your Web site and the behavior of your visitors. For these charts, you must set up scenarios containing elements that record the information you want. However, your application developers can also set up read-only charts, which display existing repository data without requiring you to record any new information. These charts show you the current values of specific repository properties and essentially act as repository status reports.
For example, your profile repository contains the e-mail addresses of the Web site visitors who have registered at your site. If you wanted to see a list of these addresses, you could set up a read-only chart that displays the current values of the Email Address property in the profile repository. As the chart is displaying data that is already in the repository, setting up a scenario to track the information is unnecessary.
To identify which of the standard charts require you to set up scenarios and which do not, display their datasets in the Dynamo Control Center (Analysis > Datasets) and look for the text "Context: read-only" in the dataset definition. The following figure shows an example:
This dataset is read only, and therefore any chart that uses it does not require a corresponding recorder scenario. In fact, read-only datasets do not appear as options in the scenario editor.
|
|