Cognos Analytics with Watson

 View Only

Enhanced data security, query performance, and modeling capabilities in Cognos Analytics 11.2.3

By Zachary Taylor posted Tue August 30, 2022 03:27 PM

  
As a product team, we enjoy seeing new capabilities sent out the door and into the hands of the user community.  In our latest release, Cognos Analytics 11.2.3, we're happy to announce integration with MS Teams, enhancements to the Cognos Analytics Assistant, forecasting with multiple parameters, multi-directional cascading filters, and much more.  In the areas of data access and data modeling specifically, we have lot of news as well. Make sure you don't miss the updates listed below!  

Data Access Updates: 
 

Support for new releases from Amazon, ORACLE, SAP, IBM, Google etc.   
As the data sources you use with Cognos Analytics are updated with new releases or features, we test them for use with Cognos Analytics so you can take advantage of new database tools and keep working without interruption. 

N
ew data source connections in CAoC On Demand
Teradata & Vertica have been supported data sources in Cognos Analytics on-premises and on Cognos Analytics on Cloud Hosted for some time - these are now available to be used in the Cognos Analytics on Cloud on Demand accounts. More data sources available means more analytics use cases can be addressed in your organization. 

Adding support for Single
StoreDB native driver
Cognos Analytics users have also been able to connect to SingleStoreDB for some time - (it used to be called MemSQL). We have been working with the SingleStoreDB team who have released a new native driver that can be used in place of the MariaDB or MySQL drivers.  The native driver can open doors for possible performance improvements, fixing compatibility issues, and providing the opportunity to add new SingleStoreDB specific features in the future.

Query engine performance: 
When you run a report or open a dashboard visualization, Cognos Analytics does work for you behind the scenes to serve up the data from your source system.  We write SQL for you so you don't have to, and we translate user gestures into planned queries that attempt to bring back data in an efficient manner to get you results faster. We support live query of your database servers and push the work for joins, transformations, and other operations to the data server when possible to reduce data movement and avoid incurring local data processing.  With various data source vendors, we make improvements over time to better utilize the functions they provide to improve the SQL that the Cognos Analytics query engine will attempt to push to the source.  These changes can help reduce query decomposition, local processing, and performance overhead.

Performance when using relative date feature
Queries generated for Relative Date measures defined in data modules are now optimized by automatically injecting an optimal set of predicates into the WHERE-clause of the generated SQL. These restrict the rows that are scanned to the minimum rows required by the Relative Date measures. This means you can get faster response times because you're processing less data.

Enhanced security options with JWT
Some teams will connect their applications to a data source and authenticate with a service account - which means all users of the app access the data from the same account.  However, many data source vendors are adding the capability to authenticate by receiving a token so each user can authenticate into the data source individually.  This allows for security to be managed for the data at the database level and allows flexibility for monitoring who accesses the data
Cognos Analytics 11.2.3 adds JWT support for Amazon Redshift with AzureAD JWT and Trino - these two are in addition to the other sources which already allow for similar functionality such as Azure SQL and Synapse, AutoRest, Exasol, Teradata, and Google BigQuery.  (Note: always check the data source vendor documentation for which version is required and see the link below title "Connections that support JWT authentication" for how to edit the JDBC connection strings when using this capability.) 

Data Modeling Updates: 


Descriptions for modeling objects: 
Modeling objects in Framework Manager have a field called Description. We've now added a similar field for modeling objects in Data Modules as well.  Unlike the Comments field, the contents of the Description field show up for Reporting/Dashboarding authors to view when authoring content. This can help the authors to understand more about the data they are working with. 
    • Descriptions – visible to report/dashboard authors
    • Comments – visible to modelers only 
    • Screen Tips – shows on hover for report/dashboard authors

Alias and shortcut tables
We are now releasing the capability to create both alias and shortcut tables in data modulesThis is a powerful capability similar to the alias and shortcut tables in Framework Manager. This functionality comes in handy when you need to join information like time dimension to multiple tables  – e.g. ship date and order date.  In the new data modules capability, you can not only create these tables, but also augment and extend them beyond what was possible in Framework Manager. 
0 comments
37 views

Permalink