...
This release includes the following no-Spark enhancements:
Support for pushing SQL expression filters on dimension columns to the MDX engine. Previously, filters based on SQL expressions could not be pushed to the Query Engine (QE) via the MDX flowIn SQL querying, Kyvos' calculated expressions, when used with a dimension filter, will now be processed through the distributed Query Engine, leading to performance enhancement.
Pruning is now supported for advanced filtering scenarios, enhancing query performance optimization.
NOTE: To enable advanced pruning, set the kyvos.query.enable.advancedPruning property to true. By default, this property is set to false.The enhanced semantic model process validates primary key constraints on dimension data values. If the primary key column contains duplicate values, the semantic model process will fail.
...
Auto Cache Strategy support for SQL Generated through Kyvos Dialogues
View Audit Logs of AI settings and User-based license information
Configure Natural Language Summary in Kyvos for a particular semantic model, which will be used during AI chat sessions and queries in Kyvos Dialogues.
Support for Auth type as 'Client ID' for 'while creating Azure OpenAI ' LLM connection for Azure deployments
Internationalization and localization Enhancements
...
Users can now add calculated measures to the semantic model, ensuring optimal query performance without any impact.
Optimized MDX queries with now support dimensions on both axes, enhancing querying query performance and efficiency.
Enhanced the performance of queries using the ‘RESPONSESUBSET’ function, leading to faster query execution and improved efficiency.
Support of Cross Join in SQL queries
...
This release comes with the following API enhancementsnew property:
Connection Property: kyvos.publish.error.data: Enables the publishing of erroneous data in the error messages and system logs.
...