Skip to main content

InterSystems IRIS Adaptive Analytics

Overview of InterSystems IRIS Adaptive Analytics

InterSystems IRIS® Adaptive Analytics is an optional extension that provides a business-oriented, virtual data model layer between InterSystems IRIS and popular Business Intelligence (BI) and Artificial Intelligence (AI) client tools. It includes an intuitive user interface for developing a data model in the form of virtual cubes where data can be organized, calculated measures consistently defined, and data fields clearly named. By having a centralized common data model, enterprises solve the problem of differing definitions and calculations to provide their end users with one consistent view of business metrics and data characterization.

Adaptive Analytics provides the following key features:

  • The Adaptive Analytics modeler, which makes data accessible for business users without exposing complex data structures, tables, or relationships.

  • Publication of data model changes as virtual cubes, preventing disruption caused by lengthy rebuilds.

  • Unified access to an online analytical processing (OLAP) model via the BI tool of your choice.

  • Live connectivity to all data stored within InterSystems IRIS.

  • Automated data structure aggregation for frequently-used queries.

Adaptive Analytics is powered by AtScale®, a product of AtScale, Inc. For documentation of AtScale functionality, consult their official documentationOpens in a new tab. Note that only users licensed for InterSystems IRIS Adaptive Analytics can view this documentation.

Configuring InterSystems IRIS for Adaptive Analytics

The configuration instructions provided here will assume that you have already successfully installed InterSystems IRIS and configured your primary namespace and databases. If you have not done so, consult the instructions in the InterSystems IRIS Installation GuideOpens in a new tab and the Create/Modify a NamespaceOpens in a new tab section of the System Administration Guide.

In order to leverage Adaptive Analytics User-Defined Aggregate Functions (UDAF), you must first import and register the UDAF class file available from the WRCOpens in a new tab with the following procedure:

  1. Download the AdaptiveUDAF.xml file from the Components area of WRC Software Distribution.

  2. On your Adaptive Analytics instance, log in to the Management Portal as a user with administrative privileges and ensure you are in your Adaptive Analytics namespace.

  3. Navigate to System Explorer > Classes and click Go.

  4. Click Import. In the modal window, select the My Local Machine option for the The import file resides on field and select the AdaptiveUDAF.xml file.

  5. Check the Compile imported items box and set the Compile flags to cuk. Click Import.

  6. Navigate to System Explorer > SQL. Execute the following command:

    CALL AtScaleUDAF.Register()
  7. To verify your configuration, navigate to System Explorer > SQL and execute the following query:

    SELECT ATSCALE_HONEYBEE_VERSION()

The configuration thus far will result in a system where aggregates are stored in the same database as source data. InterSystems strongly recommends separating aggregates so that they can be managed and assessed without needing to manually filter them from a more general data set. To separate aggregates from source data, perform the following steps:

  1. Log into a Terminal session as a user with administrative privileges. Ensure you are in your Adaptive Analytics namespace.

  2. Call the following command to configure your aggregate database and global mappings:

     write ##class(AtScaleUDAF.Utils).CreateDatabase("/<instancePath>/mgr/AtScale/")

    Where <instancePath> is the full path to your Adaptive Analytics instance.

Configuring AtScale for Adaptive Analytics

After configuring InterSystems IRIS, additional configuration is required for AtScale itself. This procedure must come after the configuration of InterSystems IRIS, and assumes that AtScale has already been installed. If you have not configured InterSystems IRIS, refer to the previous section for instructions. If you have not installed AtScale, refer to Installing Stand-Alone AtScaleOpens in a new tab in the AtScale Documentation.

After you have configured InterSystems IRIS and installed AtScale, configure AtScale according to the procedure described in Adding InterSystems IRIS Data WarehousesOpens in a new tab.

Exporting InterSystems IRIS Business Intelligence Cubes to AtScale

Adaptive Analytics supports the export of Business Intelligence cubes to Adaptive Analytics cubes. The following procedure for cube export assumes that you have already completed all of the configuration described in the previous section. This process does not in any way alter the source Business Intelligence cube.

  1. Open a Terminal session on the InterSystems IRIS instance from which you wish to export a cube.

  2. Ensure you are in the namespace of the cube you wish to export.

  3. Execute the following commands:

     set cube = "CUBENAME"
     set caltab = "CALENDARTABLE"
     set file = "PATH/EXPORTFILENAME"

    Where CUBENAME is the IRIS BI cube name, CALENDARTABLE is the name of your optional calendar table for managing dates, and PATH/EXPORTFILENAME is the target path of the cube export. The following example places an output file with the cubename and datetime stamp:

     set cube = "HoleFoods"
     set caltab = "MyCalendarTable"
     set file = "/tmp/ cubeExport "_cube_" "_$TRANSLATE($ZDT($h,3)," :","--")_".json"
     set sc=##class(%DeepSee.Utils).%AtScaleExportCube(cube,file,caltab)
    

    If no calendar file is used, simply omit the caltab argument from the final command.

    Execute the following command to verify that the cube was exported:

     write sc

    If the export was successful, this command will return the full file path of the resulting .json file. You will need this path for a later step. You may view this file in your text editor or IDE of choice.

  4. Log into your AtScale Design Center and, on the home page, select the Import from InterSystems IRIS Quick Start option.

  5. In the modal window, click Browse, navigate to the .json cube file observed previously, and open it. Back in the primary modal window, click Next.

  6. From the Data Warehouse and Schema dropdown menus, select the name of the data warehouse you have configured for this project and the appropriate schema respectively. Click Next to perform the import.

  7. In the Review Import screen, click Download Report. This report provides details of import performance and recommendations for adjustments to models, if necessary.

  8. You may receive warnings on this screen concerning incompatibilities between the structure of the export .json file and Adaptive Analytics's expected data model. These chiefly concern IRIS BI calculated measures and drill-throughs due to use of proprietary MDX operations in IRIS BI. InterSystems recommends manually reviewing all calculated measures and drill-throughs and adjust them as needed to ensure compliance with Adaptive Analytics's data model. This review process will be outlined in a later step. Click Next.

    Additionally, cubes based on data connectors and cube relationships do not export in this process.

  9. Set design-time and run-time permissions according to your needs and click Next to finalize the import process.

  10. Review your calculated measures and drill-throughs as follows:

    1. In the AtScale Design Center, navigate to your Adaptive Analytics project.

    2. Click on your imported cube to open it in the Main Canvas.

    3. Click on the Calculated Measures tool on the Main Canvas.

    4. For each Calculated Measure, click Edit and, in the modal window, click Test MDX. Consult the displayed error messages to determine appropriate adjustments to each MDX definition.

You may now review your imported cube as you would review any other Adaptive Analytics model.

Integrating Adaptive Analytics with InterSystems Reports

Adaptive Analytics integrates with InterSystems Reports, allowing users to share a common data model—virtual cubes—with other Business Intelligence tools such as Microsoft PowerBI and Tableau. Users can thereby leverage common calculations and definitions across their tool kits.

The following procedure for report generation assumes that you have already completed all the configuration described earlier in this guide, as well as the configuration described in both “InterSystems Reports DesignerOpens in a new tab” and “InterSystems Reports Server with InterSystems IRISOpens in a new tab”.

  1. Configure the Report Designer to connect to Adaptive Analytics with the Hive Driver:

    1. Launch the Report Designer and click on Hive.

    2. In the Create Connection to Hive modal window, select New Catalog. You may set Directory to a path of your choice; select the catalog you used to create the Dynamic Connection earlier for the Catalog field and click OK.

    3. In the Get JDBC Connection Information modal window, click the Driver checkbox and enter org.apache.hive.JDBC.HiveDriver in the adjacent field. Set the URL to that of your Adaptive Analytics instance, and the User and Password fields to match the administrative user of your Adaptive Analytics instance.

    4. Click More Options. In the modal window, under the Qualifier tab, select 2-Part Names from the Name Pattern pane and User Defined from the Quote Qualifier pane, accepting the provided value of ". Click OK.

    5. Once the Catalog has been generated, edit the Push Down Group Query field to true. You may now publish reports based on this catalog as normal with InterSystems Reports.

  2. Add a Report User to Adaptive Analytics:

    1. Log into your AtScale Design Center, navigate to Security and click Create User.

    2. Enter the desired information for your InterSystems Reports user and, in the Roles pane, select Runtime Query User. Click Create User.

  3. Ensure the Reports Server is configured:

    1. On your InterSystems Reports Server instance, log into the Management Portal as a user with administrative privileges.

    2. Navigate to System Administration > Configuration > InterSystems Reports.

    3. In the table of server definitions, click Configure on the row of the Reports server you wish to connect to Adaptive Analytics. If this server does not exist, follow the configuration instructions in “InterSystems Reports Server with InterSystems IRISOpens in a new tab” to create it.

  4. Configure the Reports Server to access Adaptive Analytics:

    1. In the icon menu, click My Folder.

    2. Navigate to Administration > Connections > Dynamic Connections and click New Connection.

    3. From the Catalog dropdown menu, select the catalog file you wish to use; if it does not yet exist, create it following the instructions in “InterSystems Reports DesignerOpens in a new tab”. Check that the automatically filled fields have the expected values.

    4. Click Add Database User Mapping and, in the autogenerated profile, double-click on the Database User field to edit the name. Replace it with the name of the user you created earlier in the Design Center. Accept the other defaults and click OK..

    5. In the Dynamic Connections list, click on the newly created connection.

    6. In the Properties list, double-click on the name of the connection to edit it to your liking.

    7. Navigate to Resources and run the desired report.

FeedbackOpens in a new tab