Managing models


BMC AMI Ops Insightcompares models, which are based on normal behavior in your environment, to the current behavior to see if you are experiencing problems. You can create models for LPARs and Db2 Subsystems. You can also create Db2 Data Sharing Group models by merging Db2 Subsystem models that belong to the same Db2 Data Sharing Group.

All actions support multiple selections, except for Create a model, Merge models, Copy & Create model, Edit Model Sensitivity, and Download a model options.

To access the Model Management page, on the title bar, click Administration Gear Icon.PNG

image-2024-10-14_9-47-31.png

The Model Management page includes the following tabs and options:

Option

Description

New Model

Creates new models

All Models

Displays a list of all the models that are currently available for use

LPAR

Displays the LPAR models

Db2 Subsystem

Displays the Db2 Subsystem models

Db2 Data Sharing Group

Displays the Db2 Data Sharing Group models

Model Requests

BMC.AMIOPS.SPE2501

Displays the scheduled model requests

Archived

Displays the models that are no longer available but are kept for reference

Failed Requests

Displays the models that could not be created due to an error

Show models that require retraining

 

Displays the models that require retraining

Switch theimage-2023-8-31_14-2-37.pngtoggle button to see the list of models that require retraining:

image-2024-1-17_12-37-0.png

Model Details

Displays the information related to the model

To see more details click on model itself.

image-2023-8-31_13-56-33.png

You can sort the models by:

  • Model name
  • Label
  • Created Date
  • Created By
  • Version

You can also sort the ls in the ascending or descending order.

Model Health

 

Displays the model health in graph format

You can  maximize the graph by clicking image-2023-9-6_20-17-48.png.

image-2023-9-6_20-17-17.png

To schedule model health evaluation, see Scheduling-model-health-evaluation.

Type

Provides details of model types

You can sort the models by field value and ascending or descending order.

Scoring Sensitivity

Describes the sensitivity level of the scoring engine

You can set the sensitivity levels globally or for each model between 1 to 100 percent. A higher model sensitivity detects more potential events; a lower level detects fewer potential events.

For more information about how to change the sensitivity level of each model, see To change the sensitivity level of a model.

Status

Describes the status of the models

The status of the models can be one of the following:

  • Availableimage-2023-9-6_20-7-27.png
  • Selectedimage-2023-9-6_20-6-19.png

A image-2023-9-6_20-13-12.pngnext to a model indicates that it is being used for Realtime scoring currently.

Actions menu (for multiple models)

Actions menu when selecting multiple models

The Actions menu for multiple models is available only when you select more than one model.

image-2024-1-17_12-55-29.png

Actions menu (for a single model)image-2023-8-31_13-3-51.png

Actions menu for a single model

You can perform the following actions:

  • Load model (Select)—Model is used for scoring.
    Only one model can be selected for each Db2 Subsystem, LPAR, or Db2 Data Sharing Group.
  • Unload model (De-select)—Model is not used for scoring.
  • Load (Select) for restart—Model is used for scoring when the BMC AMI Manager restarts.
  • Unload (De-select for restart)—Model is not automatically selected for scoring when the BMC AMI Manager restarts.
  • Copy & Create model—Create a new model based on existing model details.
  • You can view and modify the details.
  • Download model—Download a model to your PC to use for BMC AMI Ops Insight instance on a Docker Desktop.
  • For more details, see Running-individual-playback-sessions-in-a-Docker-container.
  • Edit Model sensitivity—Change scoring sensitivity for a specific model.
  • This overrides the default sensitivity for the model, as defined in Setting-the-global-sensitivity-level.
  • Archive—Move model to Archived tan ab.
  • In archived state, you can delete and reinstate a model.  
  • BMC.AMIOPS.SPE2501(Only available for scheduled models) Reschedule—Reschedule model creation.
  • You can reschedule the date and time. You can also cancel a scheduled model.

Model restart indicator image-2023-9-5_14-2-29.png

Indicates that a model is selected for restart

A image-2023-9-5_14-2-29.png next to a model indicates that the model will be selected when you restart the BMC AMI Manager.

image-2024-1-17_13-36-24.png

To create a new model

Use the following procedure to create one or more models to identify events:

  1. On the title bar, click Administration Gear Icon.PNG
  2. On the Model Management page, click New Model
    If you are an advanced user, you can use the Expert Mode to quickly create a model.
  3. In step 1 of 4, under Create Models, select one of the following options, then click Next:
    • Db2 Subsystem Model 
    • LPAR Model 
  4. In step 2 of 4, enter the following details:
    image-2025-10-01-11-12-38-112.png

    Fields

    Description

    Wildcard Search

      

    Uses wildcards to search a data set name

    Select the Wildcard Search checkbox and enter the wildcard input to search for a data set.

    The allowed wildcard characters are %, *, and *.

    For example: 

    • HLQ.MLQ.DATA%
    • HLQ.MLQ.*.DATA
    • HLQ.MLQ.**

    Data Set Name

    Name of the historical SMF data set that contains the source data

    In the Data Set Name field, enter one or more data set names.

    The data set must follow TSO data set naming conventions. You can use up to 500 data sets for a single model. Enter each data set on a separate line.

    Select an option

    (BMC.AMIOPS.SPE2510) Scan data sets or add filters manually
    You can use one of the following options:

    Sort SMF 

    (Recommended) Use BMC's sorting algorithm to sort the model data by date and time, to minimize resource consumption. Unless you have a specific issue that prevents you from using BMC's sorting algorithm, leave this selected.

    Warning

    Important

    If you select Sort SMF, the Pacing Record Count option is disabled. Before turning off Sort SMF, make sure the data is sorted based on date and time.

    We recommend that you enable Sort SMF.

  5. Click Next.
  6. In step 3 of 4, enter the following details based on your selection in the previous step (Add Manually or Scan & Extract), then click Next:
    • FieldDescription
      Label/Short Description

      Description for the model

      Enter a label or short description. This description is displayed in the list of models. You can use it as an alias for the model name to allow you to easily identify the model and to differentiate it from your other models.

      Output Folder

      USS subdirectory path to store the model bundles or playback file

      The product searches for this folder within the <installedDirectory>/aoidata/aoiinst/dat folder in your product installation. If the folder doesn't exist, the product creates it. Create a path that is compliant with USS naming conventions. You must have write access for the folder and sufficient space to store the data.

      Schedule model creation for later

      Creates a model at a scheduled future date and time

      Follow these steps:

      1. Under Schedule model creation for later option, click Yes.
        The default is No.
      2. In the date and time field, enter a date and time (based on BMC AMI Manager Time Zone), click OK.
      Date Range of Model Data

      Start date/time and end date/time for the historical SMF data used in the model

      The date range can be smaller than the full date range covered by the data in the data set. Dates and times follow local date/time of the data. 

      For models based on SMF 100 records, we recommend using at least a month of data.

      For models based on SMF 70-79 records, we recommend using at least 8 weeks of data.

      Warning

      Important

      To successfully create a Db2 Data sharing group model, select a date range that is included in all Db2 subsystem models that you include.

      Error
      Warning

      Make sure that the data you are using doesn't include any abnormal behavior. If it does, the model is skewed, and the product cannot provide accurate projections.

      Add Manually

      (BMC.AMIOPS.SPE2510)

       

      Create Db2 Subsystem Models (using SMF100 data)

      Creates a model for Db2 subsystems

      Enter one of the following options:

      • Filter by Db2 Data Sharing Group: Creates models for all Db2 subsystems in the Db2 data sharing group.
        A Db2 data sharing group name can be up to 8 characters long and must match the QWHADSGN field in the SMF data.
      • Filter by Db2 Subsystem: Creates a model only for individual Db2 subsystems or stand-alone Db2 subsystems.
        A Db2 subsystem name can be up to 4 characters long and must match the QWHAMEMN field in the SMF data.

      You can filter multiple Db2 Subsystems or Db2 Data Sharing Groups by using a comma separator.

      Create LPAR Models (using SMF 70-79 data) 

      Creates a model for LPAR

      Enter one of the following options:

      • Filter by Sysplex: Creates models for all LPARs in the sysplex
        Enter a sysplex name. A sysplex name can be up to 8 characters long and must match the SMF70XNM field in the SMF data.
      • Filter by LPAR: Creates a model only for the individual LPAR
        Enter an LPAR name. An LPAR name can be up to 8 characters long and must match the SMF70SNM field in the SMF data.

      You can filter multiple LPARs or Sysplexes by using a comma separator.' 

      (Default) Scan & Extract

      (BMC.AMIOPS.SPE2510)

      Select Db2 Subsystems to create a model

      Select one or more Db2 subsystems
      image (4).png

      If the selected Db2 subsystems don't have adequate data, you get a warning message.
      You can enable Db2s with adequate data option to ensure that only Db2 subsystems with adequate data are used.
      The Dataset view displays the data set names and the included Db2 subsystems. 

      Create an LPAR model 

      Select one or more LPARs
      ​​​​​​If the selected LPARs don't have adequate data, you get a warning.
      You can enable the LPARs with adequate data option to ensure that only LPARS with adequate data are used.
      The Dataset view provides the data set names and included LPARs. 

  1. In step 4 of 4, review the entries and click Submit.
  2. The process for creating a new Db2 subsystem model starts and is reflected on the progress bar on the Model Management page. The process includes the following steps:
    • Data Ingestion—Displays the number of records processed to create a model.
    • Data Training—Displays the number of records processed to train the model.
      The data ingestion is completed first and then the data training process starts. The status displays In Progress until the process is complete.
    • Data Evaluation—Evaluates the model health immediately after model creation if sufficient data is available.
      For Db2 subsystem models, you need at least seven weeks of data, where six weeks of data are for training and the last week's data is for evaluation. 
      For LPAR models, you need at least 11 weeks of data, where 10 weeks are used for training and the latest week's data is used for evaluation.

      Model_mgt.png

      When the training process is complete, the new model is listed under the model tab from which it was created. When creating Db2 subsystems models, if you filter by Db2 Data Sharing group, a separate model entry is added for each subsystem in the Db2 data sharing group. 

      Success

      Tip

      If the training request is stuck in progress, and no records are ingested, see Troubleshooting.

      Click on the model name to see the model details.
      If, while creating a new model, you exit the process before submitting it, a draft is saved. Next time you create a new model, you can either continue where you left off or start a new model creation process. Similarly, when you are copying a model and exit without submitting, the last draft is saved. 

To merge existing Db2 Subsystem Models to create a new Db2 Data Sharing Group

To build a Db2 Data Sharing Group model by merging existing Db2 subsystem models to identify events at the Db2 Data Sharing Group:

  1. On the title bar, click Administration Gear Icon.PNG 
  2. On the Model Management page, click New Model.
  3. In step 1 of 4, under Merge Existing Db2 Subsystem Models, select Db2 Data Sharing Group Model and then click Next.
  4. In step 2 of 4, under Db2 Data Sharing Group, select a group and enter the following details:
    • Label / Short Description—Enter a description for the model. This description is displayed on the list of models to help you identify your model. You can use an alias for the model name to allow you to easily identify the model and to differentiate it from other models.
    • Output Folders—Enter the location to store the model bundles. This path is created in the <installed directory>/aoidata/aoiinst/dat folder in your installation folder. You must have write access for this path and sufficient space to store the data.
      1755149633024-782.png
  5. BMC.AMIOPS.SPE2501To schedule a model creation at a later time, under Schedule model creation for later option, click Yes. The default is No.
  6. In step 3 of 4, select the models that you want to merge, and click Next.
    When selecting models to merge:
    • (Required) Select only one model for each Db2 subsystem.
    • (Required) The selected models have the same build number.
    • (Recommended) The selected models have the same date and time range.
      1755149767481-513.png

      If you select models with different dates or times, you get a message stating that the dates and times of the selected models do not match.
  7. To select the recommended models, turn on Select Recommended Models.
    All the recommended models are selected. The recommended models have a image-2023-9-21_19-32-21.pngbefore it. 
  8. In step 4 of 4, review the entries and click Submit.
    The new model is added to the All Models list.

To create a model using expert mode

 If you are an advanced user, you can use the Expert Mode to create the Db2 subsystem and LPAR models from a single screen and the same training request. 

  1. On the New Model window, click Expert Mode
    image-2024-10-11_13-56-19.png

    Warning

    Important

    You can create models for Db2 Subsystem and LPAR models in the same training request.

  2. To enter the details, use the steps provided in To create new Db2 Subsystem or LPAR model section.
  3. Click Submit to create a model.

To change the sensitivity level of a model

Warning

Important

  • When you change the sensitivity level, all active events are closed, and the scoring is restarted to generate new events based on the sensitivity level you selected.
  • Models for which you have set a global sensitivity level are marked by an asterisk ( * ).
  • If you change the sensitivity level of a single model, it supersedes the global sensitivity level.
  1. On the Model Management page, clickimage-2023-9-22_13-49-56.pngnext to a model, and select Edit Model Sensitivity.
  2. In the Edit Model Sensitivity window, select the sensitivity level.
    The sensitivity level for that model is updated and displayed under the Model Sensitivity column.
    For information about global sensitivity, see Setting the global sensitivity level.

To copy and create a model

Use this procedure to create a new model by copying an existing model. This allows you to create a model with existing data.

  1. On the title bar, click Administration Gear Icon.PNG
  2. Clickimage2022-8-10_13-34-18.pngnext to the model that you want to copy and select Copy & Create model.
    The Create & Create Model window is displayed with entries from the original model.
  3. Modify the fields as required and click Submit.

To select models for scoring

To select BMC AMI Ops Insight model for scoring and comparing with your system's current behavior.

  1. On the title bar, click Administration Gear Icon.PNG
  2. Perform one of the following steps: 
    • To select a single model, clickimage2022-8-10_13-34-18.pngto open the Actions menu.
    • To select multiple models, select more than one model to open the Actions menu at the top.

      Warning

      Important

      Loading multiple models simultaneously may take a long time.

  3. Perform one of the following steps:
    • To select the models immediately, select Load Model (Select) .
      The Status of models that are selected changes from image-2023-8-22_10-48-30.pngto image-2023-8-22_10-48-1.png

      Warning

      Important

      Only one model can be selected for each subsystem at any one time.

    • To select the models when the BMC AMI Ops Insight is restarted, select Load (Select) for restart.

      Warning

      Important

      Models that are set to load on restart are indicated by a star icon. If you want to cancel the load on restart, select Disable - Load for restart from the Actions menu.

      Models that are currently being used for scoring are marked with a Lightning_Icon.png icon.

To deselect models for scoring

Use this procedure to deselect a model that BMC AMI Ops Insight has been using.

Warning

Important

You cannot deselect models that are currently being used for scoring. You have to first stop either Realtime scoring or Playback run.

  1. On the title bar, click Administration Gear Icon.PNG.
  2. Do one of the following: 
    • To deselect a single model, clickimage2022-8-10_13-34-18.pngto open the Actions menu and deselect.
    • To deselect multiple models, select the check box next to the model names that you want to deselect and click the Actions menu at the top.
  3. Select Unload model (De-select).

The Status of models that are deselected changes from Selected to Available.

To reschedule model creation

BMC.AMIOPS.SPE2501

  1. On the Model Requests tab, select the model that you want to reschedule and clickimage2022-8-10_13-34-18.pngnext to it.
  2. Click Reschedule, modify the timings, and click Save.
    To cancel a scheduled model creation, click Cancel.

To archive models

If you don't want to use a model, but you want to keep it for later reference or later use, you can move it under archive.

Warning

Important

You can only archive models that are in image-2023-8-22_10-48-30.png status.

  1. On the title bar, click Administration Gear Icon.PNG
  2. Perform one of the following steps: 
    • To archive a single model, clickimage2022-8-10_13-34-18.pngto open the Actions menu and archive.
    • To archive multiple models, select the check box next to the model names that you want to archive and click the Actions menu at the top.
  3. Select Archive.
Warning

Important

To reinstate archived models, go to the Archive tab and clickimage2022-8-10_13-34-18.pngto open the Actions menu.

Select Reinstate.

To download a model

  1. On the title bar, click AdministrationGear Icon.PNG.
    By default, Model Management window is displayed.
  2. To download a model:
    1. On the All Models tab, select a model. 
    2. Clickimage2022-8-10_13-34-18.png, and then click Download model.
      The model is downloaded to your device with .mlb extension.  

To use the downloaded model in a BMC AMI Ops Insight instance on Docker Desktop, see Running-individual-playback-sessions-in-a-Docker-container.

When to regenerate models

In the Model Health column, you can see the evaluation of the health of your models.

image-2023-9-7_9-49-11.png

Evaluation is indicated as Poor, Good, and Excellent

(Click on the graph to see the details of the evaluations. This can show you if the evaluation indicates a trend.
Hover over a point for more information.

image-2023-9-7_9-43-2.png

If you see your models health is trending towards poor, we recommend that you generate a new model. As a general rule of thumb, models are likely to deteriorate after about three months. Results may vary according to the activity levels on your system.

For more information, see Scheduling-model-health-evaluation.

 

Tip: For faster searching, add an asterisk to the end of your partial query. Example: cert*

Administering