You are here: Decision Strategy Manager > Services > Adaptive Decision Manager service > Configuring the Adaptive Decision Managemer service

Configuring the Adaptive Decision Manager service

Use the Adaptive Decision Manager (ADM) service is to predict customer behavior. The ADM service creates predictive models and updates them by using self-learning as new responses come in. You enable the ADM service by adding one or more Pega Platform nodes. When configured for the ADM service, each node becomes a worker node that processes model updates.

  1. Click Designer Studio > Decisioning > Infrastructure > Services > Adaptive Decision Manager.
  2. Add a node:
    1. Click Add node.
    2. In the window that is displayed, select the check box for the decision data node on which you want to add the ADM service.
    3. Click Submit.
    4. Refresh the landing page until the status of the node changes to NORMAL.
    5. Repeat this step to add multiple Pega Platform nodes to the ADM service.

    Note: You need at least one decision data store node to run the ADM service. If you add an ADM node and no nodes are on the Decision Data Store tab of the Services landing page, a node is automatically added on that tab to support running the ADM service.

  3. Click Edit settings to configure the ADM default settings:
    ClosedData mart

    You can capture the ADM historical data for reporting purposes by using the ADM Data mart. The ADM Data mart is populated by periodically triggering the ADMSnapshot agent that runs the activity (pzGetAllModelDetails) that captures the state of models, predictors, and predictor binning. This activity writes that information to a table using the Data-Decision-ADM-ModelSnapshot and Data-Decision-ADM-PredictiveBinningSnapshot classes.

    1. Configure the adaptive model data storage preferences by selecting one of the following options:
      • with every update – Keeps all snapshots of ADM models. This option is required if you use model snapshots for time-based reporting (for example, trend detection). If this option is enabled over a prolonged time period, it can cause database space issues, depending on the size of your database.
      • with most recent update – Keeps only the most recent snapshot of ADM models, for example, if you want to run a report definition of only the most current version of an ADM model.
    2. Configure the predictor binning storage preferences by selecting one of the following options:
      • with most recent update – Keeps only the most recent snapshot of the predictor binning state. This option is useful if you want to analyze only the most current status of predictor binning (for example, by using a report definition).
      • None – Disables capturing the state of predictor binning.
      • with every update- Keeps all snapshots of the predictor binning state. Use this option to analyze the changes of predictor binning over time. If this option is enabled over a prolonged time period, the increased number of predictor binning snapshots can cause database space issues because you can have multiple predictors per adaptive model.

    Note:  The first time a snapshot is taken, the ADM Data mart checks for models that were created prior to Pega 7.2 and migrates them to the current version. After a successful migration, the admmart/modelIdMigrationNeeded dynamic system setting is created and the value is set to false. You do not need to repeat the migration process for successfully migrated instances. If you want ADM to trigger this step for models that were not migrated yet, change the dynamic system setting to true.

    ClosedAdaptive models update settings

    In the Models section, enter the frequency for updating adaptive models and the threshold for triggering the memory alert. The Pega Platform keeps a local cache of scoring models. The model update frequency is implemented by periodically triggering the system pulse to retrieve model updates. By default, model updates retrieve the scoring models that are required for running the strategy and the models that are different from those in the local cache.

    Perform the following actions to configure the model update frequency:

    1. Specify the model update frequency. The default frequency is every 30 seconds.
    2. In the Thread count parameter, specify the number of threads on all nodes that are running the ADM service. The default thread count is the number of available co-processes on that node, minus one.
    3. In the Memory alert threshold parameter, specify the threshold for triggering the out of memory error. The default value is 2048 MB.
    ClosedAgent schedule

    Click Edit agent schedule to modify the default agent schedule configuration of the Pega-DecisionEngine agents:

    • UpdateAdaptiveModels – Disabled by default
    • ProcessBatchJob – Disabled by default
    • ADMSnapshot – Disabled by default
    • InitializeDSMFeatureToggles – Enabled by default
    • StartDSMServices – Enabled by default
  4. Optional:  Display the details of an ADM node by clicking the row for that node.
  5. Optional: Start, stop, or decommission the ADM service node by selecting an action from the Execute menu.
Related information

Open topic with navigation