Managing projection models

To access the Model Management page, on the title bar, click . 

BMC.AMIOPS.SPE2310 

The Model Management page includes the following tabs and options:

To create new Db2 Subsystem or LPAR model

BMC.AMIOPS.SPE2310 

The product creates new models by analyzing data from your system in a process called training. Use this procedure to set up a new training request.

1. On the title bar, click Administration . 
2. On the Model Management page, click New Model. 
   Using the New Model window, you can create new Db2 Subsystem and LPAR models, merge existing Db2 subsystem models to create a new model, or use the Expert Mode to quickly create a model.
3. Under Create Models, select Db2 Subsystem Model or LPAR Model and then click Next.
   

4. On the Data Range & Data Set tab, enter the following details:

   Fields	Description
   Wildcard Search BMC.AMIOPS.SPE2310	Uses wildcards to search a data set name Select Wildcard Search check box 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.
   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 are according to the local date/time of the data.  We recommend the following for creating models: For models based on SMF 100 records, use at least a month of data. For models based on SMF 70-79 records, use at least 8 weeks of data. Important To successfully create a Db2 Data sharing group model, you should select a date range that is included in all of Db2 subsystem models that you include. Warning Make sure that the data you are using does not include any abnormal behavior. If the data includes abnormal behavior, the model is skewed and the product cannot provide accurate projections.
   Sort SMF	(Recommended) Select this to 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. 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 recommended that you enable Sort SMF.
   Pacing Record Count	Number of records processed before there is a pause in the training process Use Pacing Record Count to control resource usage of your training request. When the number of records is processed, operations are suspended for an interval equivalent to the amount of time it took to process the previous batch of records and to write them to the output file. By default the Pacing Record Count is turned off. Important If you select Sort SMF, the Pacing Record Count option is disabled.

5. Click Next.

1. On the Filter & Details tab:
   • If you select Db2 Subsystem Model, the New Db2 Subsystem Model window is displayed.
     

   • If you select LPAR Model, the LPAR Model window is displayed.
     
     On the Filter & Details tab, enter the following details:

     Field	Description
     Label/Short Description	Description for the model Enter a label or short description. This description is displayed on the list of models to help you identify your model. You can use this 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 <installed directory>/aoidata/aoiinst/dat folder in your product installation. If the folder does not exist, the product creates it. Create a path that must follow USS naming conventions. You must have write access for the folder and sufficient space to store the data.
     (For Db2 Subsystem Model) Db2 Subsystem Models Created by using SMF100 data BMC.AMIOPS.SPE2310	Creates model for Db2 data sharing group or a single Db2 subsystem  Select one of the following and enter a name for the model: Filter by Db2 Data Sharing Group—Creates models for all Db2 subsystems in the Db2 Data Sharing Group Db2 Data Sharing Group name can be up to 8 characters long and must match the QWHADSGN field in the SMF data. Filter by Single Db2 Subsystem Name—Creates a model only for the individual Db2 subsystem or Db2 standalone Db2 Subsystem Name can be up to 4 characters long and must match the QWHAMEMN field in the SMF data.
     (For LPAR Model) LPAR Models created by using SMF 70-79 data  BMC.AMIOPS.SPE2310	Creates model for Sysplex or LPAR Select one of the following and enter a name for the model: Filter by Sysplex—Creates models for all LPARs in the Sysplex Sysplex 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 LPAR can be up to 8 characters long and must match the SMF70SNM field in the SMF data.

2. Click Next.

3. On the Review & Submit tab, check the information provided and click Submit.

   BMC.AMIOPS.SPE2310 The process for creating new Db2 Subsystem or LPAR models starts and is reflected on the progress bar on the Model Management page. The process includes:

   • 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.
     
     
     If you have a training request in progress, you cannot issue another training request. If you do, an error message is displayed.
     

     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 Db2 subsystem in the Db2 Data sharing group. When creating LPARs models, if you filter by Sysplex, a separate model entry is added for each LPAR in the Sysplex. 

     Click on the model name to see the model details.

To change the sensitivity level of a model

BMC.AMIOPS.SPE2310 

1. On the Model Management page, click next 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 Scoring Sensitivity column.
   For information about global sensitivity, see Setting-the-global-sensitivity-level.

To create a model using expert mode

BMC.AMIOPS.SPE2310  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.
   

   Important

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

2. Enter the details as explained and click Submit.

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

BMC.AMIOPS.SPE2310 

Use this procedure to merge Db2 subsystem models into a Db2 Data Sharing Group model.

1. On the title bar, click Administration .  
2. On the Model Management page, click New Model.
3. Under Merge Existing Db2 Subsystem Models, select Db2 Data Sharing Group Model and then click Next.
4. On the Select Db2 Data Sharing Group tab, select a Db2 data sharing 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.
     
5. Click Next.
6. BMC.AMIOPS.SPE2310 On the Select Models tab, select the models that you want to merge by clicking the check box next to them and click Next.
   When selecting models to merge:
   • (Required) You 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. 
     
     
     If you select models having different date or time, you get the following message:
     
     

1. To select the recommended models, turn on Select Recommended Models.
   All the recommended models are selected. The recommended models have a before it. 

2. On the Review & Submit tab, click Submit.

   The new model is added to the All Models list.

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 . 
2. Clicknext 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 . 
2. Do one of the following: 
   • To select a single model, clickto open the Actions menu.

   • To select multiple models, select more than one model to open the menu.

     Important

     Loading multiple models simultaneously may take a long time.

3. Do one of the following:
   

   • To select the models immediately, select Load Model (Select) .
     The Status of models that are selected changes from to . 

     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.

     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  icon.

To deselect models for scoring

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

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

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

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.

1. On the title bar, click Administration . 
2. Do one of the following: 
   • To archive a single model, clickto open the Actions menu.
   • To archive multiple models, select the check box next to the model names that you want to archive and click the menu at the top.
3. Select Archive.

To download a model

1. On the title bar, click Administration.
   By default, Model Management window is displayed.
2. To download a model:
   1. On the All Models tab, select a model. 
   2. Click, 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

BMC.AMIOPS.SPE2310 

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

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.

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.