IBM is thrilled to announce the general availability of IBM Cloud Pak for Data version 5.2. Built on the solid foundation of version 5.0 and 5.1, Cloud Pak for Data continues to evolve as a unified framework that lets you connect to your data, govern it, find it, and use it for analysis. This evolution also includes the decoupling of the platform as IBM Software Hub to install, manage, and monitor IBM Data and AI solutions on Red Hat OpenShift. Read more about IBM Software Hub and the enhanced add-ons in IBM Software Hub Premium blog. But rest assured that the deployment of Cloud Pak for Data services stays the same with no additional installation or upgrade steps required.

Using Cloud Pak for Data's modular, containerized framework, teams can start with the essentials and quickly expand by deploying the services they need. Whether leveraging trusted data with Match 360, accessing distributed data through Data Virtualization, enhancing governance with IBM Knowledge Catalog, or analysing data with Watson Studio customers can build a strong data foundation to accelerate AI and analytics.

Cloud Pak for Data service enhancements and updates:

IBM Knowledge Catalog introduces powerful updates to streamline governance, automate data quality, and enhance metadata management:

• Consistent Metadata Across Workspaces: Identical data assets across catalogs and projects now share a single set of properties. Update once, and changes sync everywhere—ensuring consistency and reducing manual effort.

• SQL Query Assets in Catalogs: You can now publish SQL query assets to governed catalogs. Previews respect data protection rules, enabling secure, reusable query sharing.

• Execution Windows for Metadata Jobs: Schedule metadata import jobs to run only during defined time windows. Jobs pause if they exceed the window and resume automatically later—ideal for managing system load.

• Automated Data Quality Checks: Data quality analysis is now smarter. Checks can be auto-generated from profiling results or business term constraints, then reviewed or applied directly. New checks include historical stability and referential integrity.

• Rule-Based Term Assignment: Define and upload term assignment rules via CSV. The system applies business terms to assets and columns automatically, accelerating enrichment.

• Expanded Data Source Support: Import, enrich, and assess data quality from remote file systems using FTP connections—broadening your governance reach.

• Governance Relationships with Viewer Access: Users with only Viewer permissions can now create custom relationships between governance artifacts, increasing collaboration without compromising security.

• Improved Relationship Explorer: View indirect relationships via dashed lines and choose how item names are displayed—current, original, or AI-generated.

IBM Match 360 introduces key enhancements to improve data stewardship and simplify master data configuration:

• Smarter Overlay Remediation: A redesigned comparison table and simplified task controls make it easier to review, accept, or reject incoming record updates—ensuring only accurate changes are published.

• Unified Data Type Management: Manage all data types—attributes, records, entities, hierarchies, groups, and relationships—from a single screen. Define or update types and configure matching logic for each entity in one place.

IBM Data Virtualization introduces new capabilities to enhance data protection, governance, and integration flexibility:

• Automated Data Lineage with MANTA: Track end-to-end data flow by importing metadata using MANTA Automated Data Lineage—gain visibility into data origins, transformations, and destinations.

• "Mask at Read" Semantics: Strengthen data protection by applying masking rules before query operations like predicates, JOINs, GROUP BY, and ORDER BY. This goes beyond default result-set masking for deeper security.

• Db2 Datalake Table Support: Replace legacy Hadoop table syntax with modern Db2 Datalake table semantics for Cloud Object Storage. Use CREATE and DROP DATALAKE TABLE statements for streamlined access and caching.

• Shared Properties for Catalog Governance: Ensure consistent governance across multiple catalogs by managing connected data assets with shared metadata properties.

IBM continues to enhance the AI development experience with new features in Watson Studio and Watson Machine Learning:

Watson Studio

• Document Editor for Project Lifecycle: A new built-in Document editor lets you create and manage project documentation, including README files, directly from the project Overview page—making collaboration and tracking easier.

• watsonx Asset Integration with Git: You can now import and publish watsonx™ prompt templates and vector index assets in projects that use default Git integration, streamlining asset management in Cloud Pak for Data.

Watson Machine Learning

• Support for Runtime 25.1: Deploy machine learning assets using the latest software specifications based on runtime 25.1 for improved performance and compatibility.

• Deploy Spark 3.5 Models: Models trained with Apache Spark 3.5 are now supported for deployment, expanding flexibility for big data and distributed ML workflows.

These enhancements improve metadata consistency and governance, helping teams maintain high-quality master data with less effort using IBM Knowledge Catalog. While streamlining workflows and centralized configuration give teams greater control over data quality with Match 360. New features in Data Virtualization strengthen data protection, lineage tracking, and hybrid data access - enabling more effective governance and virtualization. Watson Studio includes improved documentation tools and expanded integration options empower teams to build, manage, and deploy AI solutions more efficiently. These are main highlights for some of the Cloud Pak for Data base services, for more information about these features and other services check out What’s New in the IBM Documentation.

Check out the announcements on ibm.com:

Making data work smarter: What’s new in IBM Cloud Pak for Data 5.2

Introducing IBM Software Hub and Software Hub Premium