Product Information > JADE Object Manager Guide > Chapter 4 - JADE System Instrumentation and Diagnosis > Using the Jade Sampling Dialog
Using the Jade Sampling Dialog

The Jade Sampling dialog enables you to select the options that you require for sampling, as follows.

  1. The Active Nodes panel displays all currently active nodes on the system and is dynamically updated whenever a node begins or shuts down.

    To select the nodes that are to be sampled, simply check the appropriate check boxes. (As node sampling is disabled by default, no active nodes are checked when the JADE Sampling application in invoked.)

  2. By default, all classes in all schemas are sampled. The Jade Sampling Class Selector dialog enables you to select specific classes that are included in the sampling context, by clicking the Set Filters button.

    For details about selecting the classes that are whose statistics are sampled, see "Using the Jade Sampling Class Selector Dialog", in the following section.

  3. The Sampling group box enables you to choose between sending your output to a file on disk or through a TCP/IP connection.

    Unless you have written your own TCP/IP interface, use the default Output to file option, which is checked by default.

    The examples\sampling directory on your JADE release medium contains sample code.

    If you do not specify a file name in the File name text box, your remote node sampling is output to the jomstats.smp file in the JADE installation directory by default.

    If you want to retain an existing sampling file (for example, you have a sampling context that was output to the default file), ensure that you define a unique name for the next sampling context, as existing files are cleared if sampling is started to a file whose path and name already exist.

    To ensure that file names do not conflict, you can include one or more macro definitions, or escape characters, within the file name; for example, c:\jade\performance\stats%p%d%t.log. The macro definitions that you can use are listed in the following table.

    Definition Description Example
    %p Process id 123
    %d Date 31/07/2000
    %t Time 07:05:57
    %c Computer name statsjeken
  4. To select the detailed operations that are to be sampled, check the appropriate check boxes in the Detailed Operations group box. You can select one or both of local or remote individual requests and the cache activities that you want to sample. (For details, see the Node::beginIndividualRequestsLogging method or System::beginIndividualRequestsLogging, earlier in this chapter.)

    The Start button is enabled only when sampling has begun, to enable you to start or stop detailed operations at any time during an active sampling context. When sampling has ended, the caption of this button is displayed as Stop.

  5. To take a snapshot of statistics for local or remote activity or both local and remote activity (record types 8 and 9), check the appropriate check boxes in the Log Request Statistics group box. (For details, see Node::logRequestStatistics or System::logRequestStatistics, earlier in this chapter.)

    The Take Snapshot button is enabled only when sampling has begun, to enable you to take snapshots of log request statistics at any time during the active sampling context. When sampling has ended, the button is disabled.

  6. To take a snapshot of statistics for cache statistics or cache buffers or for both cache statistics and buffers (record types 1 and 2), check the appropriate check boxes in the Log Objects Caches group box. (For details, see Node::logObjectCaches or System::logObjectCaches, earlier in this chapter.)

    You can sample any combination of persistent, transient, or remote transient cache statistics. All buffers containing non-shared transient objects are listed when node sampling snapshots are requested.

    The Take Snapshot button is enabled only when sampling has begun, to enable you to take snapshots of log object caches at any time during the active sampling context.

    When sampling has ended, the button is disabled.

  7. The User Text and User Number text boxes enable you to specify any user information that you require, which is recorded on each record. (For details, see "System Class External Methods", earlier in this chapter.)

  8. The Extract Class Names check box, which is checked by default the first time you access the Jade Sampling dialog, enables you to extract the class names and numbers and the schema name, class name, method name, and oid of each method in all schemas to a file that defaults to jomstats.txt in the JADE installation directory; for example:

    z:\jade\bin\jomstats.txt
  9. When you have selected all of the sampling options that you require in the dialog check boxes, click the Begin button in the Sampling group box.

    The Start button in the Details Operations group box is then enabled.

  10. To start the sampling context, click the Start button in the Detailed Operations group box.

    The sampling context for your selected criteria is then started, the caption of the button in the Sampling group box changes to End, and the Take Snapshot buttons in the Log Request Statistics and Log Object Caches group boxes are enabled.