What's new in IBM Db2 AI for z/OS 1.6.x
Db2® AI for z/OS® delivers new and enhanced features in a continuous manner. Review the summary of changes in the latest release, to help you plan your installation, migration, or upgrade.
What's new in version 1.6.4?
- Provides greater insight into access path related regressions that might occur on the system,
and provides flexible options for resolving those regressions. This includes the ability to:
- Automatically resolve any access path regression that might occur according to a customized schedule.
- Automatically resolve only access path regressions that pertain to Db2ZAI access path recommendations, according to a customized schedule.
- Review access path regressions and resolve them according to a customized schedule.
What's new in version 1.6.3?
- To reduce manual intervention, Db2ZAI is adding support for the scheduling of SQL optimization processes. Schedules can be configured to run daily, weekly, or at any specific time in the future.
- Additionally, the SQL optimization feature will now support searching by client user ID and client APPLNAME, making it easier to identify and list dynamic SQL statements directly on the dashboard.
- Db2ZAI 1.6.3 also updates the metrics displayed in the DCC dashboard to provide more accurate and useful information for analyzing DCC issues.
What's new in version 1.6.2?
- The ability for users to choose the dynamic queries for Db2ZAI to optimize by providing the client user ID or the client application name (or both) under which the queries are executed.
- Improved reliability of the access path regression detection logic used by Db2ZAI through the use of standard statistical methods (standard deviation, standard error, and confidence level) for comparing sets of data.
- The ability to run assessments on Db2 subsystems that are not connected to Db2ZAI.
- The ability to select the exception types and categories included in assessments.
- The flexibility to schedule multiple recurring assessments with shorter time frames than 24 hours.
- A connections high water mark section to analyze the number of connections in use in comparison to the CONDBAT.
- The option to suppress DCC warning messages in the Db2 console.
- Introduces a simplified migration process for moving to new levels of Db2ZAI.
- For Db2ZAI 1.6.2, all connections to Db2 use a PassTicket. Passwords and any administration tasks related to password management are no longer required.
What's new in version 1.6.0?
-
Introduces the access path exploration feature. This feature leverages query execution data collected by Db2ZAI on the target connection and combines it with data collected through an exploration process, to find better performing access paths for the queries being optimized.
Optionally, the performance of these access paths can be validated during the exploration process to ensure that no performance regression occurs after they are deployed. In addition, this exploration process can predict whether an access path performance regression is likely to occur and, if so, recommend actions to take that prevent the regression from occurring.
-
Introduces enhancements to the System assessment and Performance insights functions including:
- New logic to the exception analysis to evaluate exceptions related to Db2 13 metrics.
- New System assessment trend analysis for system programmers to evaluate exceptions that occur over a long period of time.
- New interactive pivoting capabilities for custom data analysis.
-
Introduces enhancements to Distributed connections control function including:
- Ability to edit DCC recommendations across members.
- Added utilization of the new IFCID 417 trace to gather client inventory information. IFCID 417 data helps DBAs correlate data between IP addresses, user IDs, and application names.
- IBM Db2 AI for z/OS 1.6.0 provides the following usability enhancements:
- Simplifies the software stack by removing the dependency on IBM Watson® Machine Learning for z/OS and IBM Open Data Access for z/OS, to enhance the efficiency and speed of new installations, upgrades, and ongoing maintenance.
- Embeds Spark and machine learning libraries to provide a more simplified up and running experience and faster time-to-value.