Model Execution

When clicking the Run ("model.jma") Shift + F6 in OMS and the executor for a model is set to Internal or External an user interface for execution shows up. This means that the model gets loaded and is ready for execution. You can parametrize the model with data from parameter files, look at the current state of attributes, debug the model at the attribute level, and browse the output.

In this window you can finally run the model by pressing the green button for Starting/Resuming the model execution.

The execution user interface has three main panels that are accessible in tabs.

Model Parameterization (Data Tab)

The model parameterization and data editing window is shown in the Figure below. All Attributes of the model are listed here with name and value. The top right corner contains a button that shows the total number of attributes in the model and the number that is currently visible. Pressing this button will show all the attributes. You can sort the view ascending or descending with the buttons right next to it. You can also filter the current attribute view by

• Attribute name (shown in the Figure): Select the name pattern you want the view to filter by.
• Attribute type: Show all the attributes of a certain type.
• Attribute unit: Show all Attributes that have the same physical unit.
• Attribute dimension: Show all attributes that have the same dimension.
• Role Filter:Show all attributes that play the same role in a model. Default roles are for example Parameter or Variable or Dimension
• Component Filter: Show all attributes that are referenced from a certain component.

Figure: Model Data Editing

Attributes can be edited within the visible attribute table. However, if the Properties window is visible, they can be edited there as well.

Note that attributes with different types can be visible in this view, such as scalars, strings, and array attributes at the same time. Array attributes use a bracket notation that indicates row/column grouping.

Arrays can also be edited in a table. Select the array attributes you want to edit and click on the table icon in the tool bar that should become enabled. The view switches to a table that shoes the array elements in the right order.

Property Parameter Sets (PPS Files)

Property parameter sets are ASCII files containing sets of data that can be applied against a model. It can be loaded using the folder button in the tool bar of this tab.

A property parameter set file has the extension pps. An example is given below:

...
/@nac=1
/@albset_sna=0.05000000074506 
/@nlapse=3
/@radpl_lat=[38.5 38.5 38.5 38.5 38.5 38.5 38.5 38.5 38.5 ]
/@radpl_aspect=[0.0 90.0 90.0 45.0 45.0 45.0 135.0 135.0 90.0 ]
/@epan_coef=[1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 ]
..

It consists of pairs: <Mpath> = < Value>, where MPath points to the location of the attribute in the model, and Value is the text value to use.

Model Output Visualization (Output Tab)

The output tab shows the produced output after the model ran. For each execution it produces a folder, that stores the output data. Double clicking on a table will open up a graph and spreadsheet.

Figure: Model Output Files

Model Debugging (Model Tab)

The Execution runtime supports debugging at the model level. This means that the execution can be stopped and resumed before/after a component gets executed. In the Model tab you can set breakpoints on which the model should stop during execution. It is (currently) not possible to set breakpoints within components.

Figure: Model Breakpoints

Once the model is being stopped at a certain component, you can switch to the data tab to explore the current state of the model attributes. This might give you some indication about computational problems within the model. You can resume execution by clicking the green arrow again. At this time breakpoints can also be cleared if desired.