Data Designer's general purpose is to enable the creation and management of metadata. It has also been tailored to the needs of managing regulatory requirements like the
Firm Data Submission Framework.
Data Designer's current form is an Eclipse tool based on the
Eclipse Modeling Framework (EMF) and
EMF Client Platform (ECP). As such, an
Ecore model was implemented to enable the auto-generation of code and the Data Designer platform. The Ecore model drafted is a variation of the
European Data Point Model also used to model the
CRD IV / CRR framework, also known as
FinRep and CoRep.
Getting Started
 |
| Figure 1: Data Designer layout |
After Data Designer has started for the first time you will see three empty panels. The top left panel is the Model View, where all model elements belonging to a project can be created, navigated and deleted through a tree structure. The bottom left panel is the Model Repository View where existing model repositories can be accessed and new repositories added. The main, central window is the form Editor where model elements selected from the Model View are displayed in a form for editing.
Data Designer is an extensible platform built on the OSGI plug-in architecture. This means new data models can be designed and immediately used. The example we will use here is the data model designed for the Prudential Regulation Authority’s Firm Data Submission Framework.
Installing a Data Model
When first running Data Designer, you may need to install the data model plug-in if it has not already been installed. To see if it is installed:
- select Data Designer | About Data Designer in the menu
- in the pop-up, click on the 'Installation Details' button
- the FDSF data model has been installed if you see a Feature with the id 'com.jcchapman.fdsf.model.feature' in the 'Installed Software' tab
 |
| Figure 2: Installing a data model |
To install the data model:
- select Help | Install New Software
- in the 'Install' wizard, click Add
- in the 'Add Repository' dialogue, enter a name and select Local
- navigate to where you have the Data Model Feature stored and select Open
- the 'Location' in the 'Add Repository' dialogue will now show its file path so click OK
- you will now be back in the 'Install' wizard. Select the check box next to your data model and click Next
- continue to click Next through the rest of the wizard, accepting the licensing agreement, until you finally click Finish
Viewing and Editing Data
Now the data model is installed you can view and edit any data conforming to this model. If you have been provided the data in XMI (XML Metadata Interchange) format you can easily import it into Data Designer for immediate use. To do this right click in the Model View and select 'Import Project'. Navigate to where the XMI file is stored and select Open. You will be prompted for a project name. Enter the name you would like displayed in the Model View and click OK. A tree view of your data should now appear in the Model View.
 |
| Figure 3: FDSF displayed in Data Designer |
The above shows the Firm Data Submission Framework. Specifically, the model element for CP Capital Stress Testing is shown with supporting documentation listed as annotations in the Model View and the required data listed as Contexts in the Form Editor.
Everything shown is an exact reflection of the underlying data model. The risk that a software implementation does not accurately reflect a data model design has been significantly reduced.
This includes all relationships between objects in the data model. These are reflected as links between model elements. For example, a Framework consists of a set of Contexts. Clicking on the Retail Mortgage Context opens its respective form. The Model View automatically navigates the tree to the Retail Mortgage context so we can see all of the Dimensions and Measures (e.g., Comments, LTV-Arrears PIA, etc.).
 |
| Figure 4: Everything shown is a reflection of the underlying data model |
Implementation of the data model also includes validation. For example, any mandatory model elements are indicated with an asterisk and are marked with a red exclamation mark if they have not been populated. The chosen UI elements in the Form Editor are a reflection of the underlying data types. For example, Name and Definition are strings resulting in text boxes, Valid From is a date resulting in a data picker and 'Used by Data Item' is a relationship resulting in a list of hyperlinks. This automated functionality is thanks to the underlying architecture of EMF in association with ECP
 |
| Figure 5: Adding data model elements |
Additional hyperlinks can be added by simply clicking on the two icons in its right corner. Specifically, they allow one to link existing model elements as well as create and link new model elements.
Right clicking in the Model View brings up the context menu. When we right click on Data Model we see the list of possible Model Elements we can add. The menu also gives options to delete a model element, import or export parts of the data in XMI format, import a complete project, generate documentation and generate a
XML Schema Document (XSD).
If we right click on a specific element of the data model we will see only those model elements we are allowed to add. For example, for a Context in the FDSF data model we can only add an Annotation or a Data Item Relationship.
FDSF Example: Adding a Context
As an example, lets add a Context to FDSF. A Context combines dimensions and measures to create a data request. Our Context will request Drawn Balance data split by Portfolio and the percentage that they are in arrears.
First, right click on Data Model, as above, and select Context. This creates a new blank Context. The Name field has a red exclamation mark, indicating that this is a mandatory field that must be populated. As per the designed data model, Measures and Dimensions are related to a context via Data Item Relationships. This enables one to input additional details, such as the minimum of maximum occurrences of that data item or its default value for the Context. The Model View shows that we have also added an Annotation so we can enter additional details describing the purpose of the Context.
 |
| Figure 6: Adding a Context |
The next figure shows that we have named the Context 'Drawn Balance Split by Arrears'. This name is automatically reflected in the Model View, as is the title of Purpose for the Annotation, 'Portfolio' (being a pre-existing Context of a series of dimensions like Commercial Entity, Business Unit, etc.) and 'Arrears Band MIA Cumulative' being an existing Explicit Dimension of a set of bands specifying months in arrears. The figure shows that we are in the process of linking the third Data Item Relationship to the Measure called 'Drawn Balance'. To do this, we've clicked on the 'Of Type Data Item' link, which has brought up the Search Elements dialogue box. Search on Drawn Balance we are given all model elements that contain this string. So we can now select 'Drawn Balance' and the definition of our Context is complete.
 |
| Figure 7: Linking a Measure to a Context using a Data Item Relationship |
Working Sets
 |
| Figure 8: A working set is a predefined filter on the data model elements to display |
Sometimes we may only want to focus on a subset of the full list of metadata. By clicking on the down arrow in the top right corner of the Model View we can define a Working Set. A Working Set enables one to define a subset of metadata model elements that one may want to focus on. Upon definition and selection of a working set, only those model elements included in the working set will appear in the Model View.
By clicking 'Select Working Set' we are presented with the 'Select Working Set' wizard where we can click 'New'. This starts a new wizard where we select 'FDSF Working Set' and click Next to go to the 'New Working Set' dialogue. This final dialogue box allows us to select any combination of Contexts, Explicit Dimensions and Measures for our Working Set. Specifically, as shown in the figure, we can set the name of the working set, filter the list of Model Elements (e.g., in the figure below we are filtering all elements containing the word Balance) to make navigation easier and click the check boxes of those Model Elements we wish to appear in our Working Set.
 |
| Figure 9: Creating a new working set |
Generating Documentation
Data Designer also enables customisations like the ability to export different types of documentation. Here we have created the ability to export documentation in HTML format or a XML schema document, for the full data model or a specific working set. These are available via the icons in the top right of the Model View.
This
link takes you to an example of the HTML documentation export and this
link takes you to an example of the XSD export. Both have been done for a working set defined for CP Capital Stress Testing.
