Cognos Analytics

 View Only

5 Data Access and Data Modeling Updates Not to Miss in Cognos Analytics 11.2.4 (LTSR Dec 2022)

By Zachary Taylor posted Fri December 16, 2022 10:58 AM

  

In recent years, the data generated by businesses has exploded  - this explosion is not only in the volume of data available, but also in the variety of data types that can be used.  The information that is needed to answer business questions can come from many sources and may not be stored in a typical enterprise data warehouse - it might come in many different formats and reside in many locations around the organization. Some examples of the different types of data you might want to analyze are shown below:


(Examples of data source types that could be used to answer business questions)


Getting access to that data and then getting it in shape to answer business questions is something IBM Cognos Analytics can help with.
We provide tools for data access and data modeling/preparation to help derive business insights. 

Check out the overview info on data access and data modeling and keep reading the highlights below to learn examples of what’s new in data access and data modeling in the latest version (11.2.4 LTSR, Dec 2022):



Databricks Connector

It helps to have data in a structured, organized form to quickly answer business questions, but because data is being generated so quickly, it doesn’t always get cleaned up and archived into a data warehouse format right away. While some data may be archived in a structured way, other data is left in crude/raw format and placed into low cost cloud storage locations as collections of files. A technology like IBM Cloud Data Engine can be used to query files in stored in locations like IBM Cloud Object Storage.  Databricks is another technology that is used to help manage both structured and unstructured data in a data lake. IBM Cognos Analytics 11.2.4 now supports querying Databricks as a source for answering business intelligence questions.

Main menu > Manage > Data Server Connections > “Add data server”



CDATA connector

Other new sources of data for answering business questions could be business applications that people use for specific tasks in their jobs. These job activities could include managing marketing campaigns, completing outage incident reports, updating employee profile details, or updating sales lead notes. Interesting and helpful data about these tasks that could prove insightful for the business could come from sources like: 

        1. Marketo

        2. ServiceNow

        3. Jira

        4. Workday

        5. Dynamics CRM

The data in these systems may not be immediately accessible for analysis, and even if they users do connect to a REST endpoint to access this data, it may come in a JSON format that is not immediately useable for BI analysis. In Cognos Analytics, however, users can set up a connection to their REST endpoint using the Cognos Analytics provided connector called Progress DataDirect Autonomous REST API connector. This connection allows you to pull data from an endpoint and also convert the JSON format into a relational, flat style that can be joined with other sources.  And now, as of Cognos Analytics 11.2.4, users can also connect to CDATA Connect Cloud which allows you to more easily connect to a wide range of sources where business data may reside (see examples below). 



Dynamic Schemas

Sometimes an organization will have multiple versions of a particular data source - for example sample data in a test database and real data in the production database. Instead of making two separate data models, however, it could be easier to build a single data model for preparation and then simply point at the sample data or the production data for different tasks. 

In another related scenario, an organization who uses the Cognos Analytics multi-tenant capabilities may need to change which data sources is used based on which tenant is running the data model.

In Cognos Analytics 11.2.4, we introduce a new UI to allow users with these tasks. A data module is configured to look for a certain source, but you can now have that default source selection be overridden and have the model use another source. (Dynamic schemas were supported as of 11.1.7 FP1 and 11.2.0, but were implemented without a UI which has now been added.)

At runtime, a data module can be dynamically redirected to a different data servers/catalogs/schemas that has been defined.  A user can write a macro expression, for example, to choose a different data source based on some session variable like the tenantID or machine name.  This allows for new test, dev, prod planning configurations. 

For a more in depth coverage of this topic, check out the blog post by my colleague Ian Henderson here.



JWT support added for Dremio

When you connect Cognos Analytics to a data source, you have choices about how to authenticate/log in to the data source - you can attempt to connect to the source anonymously without credentials (not recommended except in special scenarios), you can have each user who needs access to the data to provide credentials to log into the data source themselves, you can store a set of shared credentials like a service account, or use an external namespace.

When using an external namespace, certain data source vendors allow for the passing of a JSON Web Token (JWT) from the OIDC authentication provider down through the data source driver to the source. This means you get the security of having each person log into the source individually but without the hassle of having to require them to type in their data source name and password every time.  This “SSO-like” experience allows for data security to be managed at the data level (in the source system) and also provides for the data source admins to view who specifically is requesting the data. 

We support a growing number of sources that allow for this type of interaction, and the latest one we’ve added in Cognos Analytics 11.2.4 is Dremio SaaS. Check out the list of data sources that support JWT authentication in Cognos here, including Dremio, Azure SQL, Snowflake, and more.



Capabilities for Data Modules, Data Sets, and Uploaded Files

In Cognos Analytics 11.2.4, we've also added a new level of granularity to the ability to control user capabilities.  You can now give a user the ability to perform certain tasks generally, but restrict these same capabilities when using specific data sources (Framework Manager packages, data modules, data sets, and uploaded files).  

For example, a user who can normally build dashboards can be prevented from creating a new dashboard, turning on the dashboard edit mode, copy/pasting widgets, saving, or relinking a dashboard when using a specific source.  You can also restrict users from creating a new data module, enriching a package, or opening a data module when using a restricted source. 

This means that you can allow users to have authoring capabilities in general, but restrict them in special cases with specific sources. (Note: these restrictions may not fully apply when the user is the author of an asset in which case they retain capabilities.)  For more information, see the documentation here




_______

The goal of Cognos Analytics is to get the right data to the right people in the best way.  These new features provide more flexibility to allow you to do just that. 

Check out the new features by accessing 11.2.4, and let us know what you think in the comments below! 


#CognosAnalyticswithWatson
4 comments
82 views

Permalink

Comments

Wed January 18, 2023 09:50 AM

Hi @Bek Ibragimov are you using the latest version of CA - 11.2.4? 

If you are and still are experiencing issues, you can submit a support ticket at https://www.ibm.com/mysupport/ 

Tue January 17, 2023 07:56 AM

I am trying to create Databricks connection but having following error: XQE-JDB-0004 A problem occurred finding the driver class "com.databricks.client.jdbc.Driver". Any ideas?

Fri December 16, 2022 04:39 PM

Excellent - glad to hear it!

Fri December 16, 2022 03:42 PM

These are all welcome changes!