Guanjun Cai, Akiko Hoshikawa, and Jason Cu

Your most valuable business data lives in Db2 for z/OS on IBM Z. But are you tapping into this exponentially growing trove of transactional, financial, and operational records for actionable insights? Are you capitalizing on the insights for business transformation and growth? Traditional SQL queries can only take you so far. They find exact matches but often miss the hidden patterns, subtle relationships, and semantic connections where meaningful intelligence resides.

IBM SQL Data Insights Pro (SQL DI Pro) changes that. This AI-powered solution brings advanced semantic search and pattern discovery directly to your Db2 data, unlocking hidden insights without the complexity, cost, or risk of moving data off the mainframe.

How does SQL DI Pro work?

SQL DI Pro integrates seamlessly into your Db2 environment through three main components working together. The model training engine, running on IFLs, builds vector embeddings from your data without impacting your mission-critical Db2 workloads. The data and query engine, embedded natively in Db2, provides data processing and semantic query services through built-in scalar functions. The client application, consisting of a web UI, RESTful APIs, and command-line tools, enables you to easily manage AI objects, train models, and run queries.

The workflow is straightforward: You connect SQL DI Pro to your Db2, create an AI object from a table or view, and enable the object for AI queries. SQL DI Pro handles data preprocessing and model training automatically. Once enabled, you can run similarity, dissimilarity, clustering, analogy, or commonality queries by using simple SQL statements. SQL DI Pro handles all the AI complexity behind the scenes. As your data changes, you can retrain models fully on the entire data or incrementally on a subset of data. To avoid production disruption, you can deploy the retrained model at your convenience.

What makes SQL DI Pro so advanced?

SQL DI Pro brings practical AI and meaningful insights to your fingertips through the following advanced features and capabilities:

• Unified analysis of structured and unstructured data. Combine traditional database columns with free-text fields for more accurate and comprehensive analysis. This unified approach extends your analytics capabilities far beyond what structured data alone can provide.

• Discovery of hidden patterns and connections. Identify similar cases, claims, transactions, or customer profiles across records in tables and views, revealing opportunities for business strategy transformation, process optimization, and operational improvement.

• Anomaly detection before it escalates. Spot unusual behaviors, outliers, and suspicious activities early. SQL DI Pro flags records that deviate from normal patterns, helping you catch fraud, identify risks, and prevent losses.

• Semantic search by meaning, not just keyword match. Find relevant records based on context and intent, even when different terminology is used. Search for "payment issues" and discover records mentioning "billing problems," "invoice disputes," or "transaction errors"—terms that traditional keyword searches would miss.

• Simplified and accelerated AI solution development. Streamline key parts of the AI workflow to reduce the time and effort required to build and deploy AI-based solutions. By connecting directly to your Db2 data and enabling AI queries, you can focus on exploring insights sooner—without the overhead of manually stitching together infrastructure or tooling. This helps you move from data to value faster, with significantly less setup and operational complexity.

• Model freshness sustained through incremental retraining. As your data evolves, update models incrementally without costly full retraining cycles, ensuring your insights remain accurate and relevant.

What sets SQL DI Pro apart?

SQL DI Pro is built natively in Db2 and optimized for IBM Z and LinuxONE, bringing AI capabilities directly to where your data resides. Using database embedding and deep self-supervised learning, it trains neural network models that understand the semantic meaning of your data—not just literal values. This enables you to discover insights that traditional queries simply cannot find, such as identifying similar records despite different terminology or detecting subtle patterns that signal fraud or operational inefficiencies.

What truly sets SQL DI Pro apart is its seamless integration with your existing infrastructure. This means that you can leverage your current Db2 and Z investments while taking full advantage of their unmatched performance, reliability, and scalability for AI workloads. Your mission-critical data stays intact, compliant, and secure on Z without the costly, risky, or time-consuming data movement off the platform. Also, you run AI-powered queries by using familiar SQL syntax, eliminating the need for specialized AI expertise and extensive model management. This makes advanced AI accessible to business analysts, database administrators, and application developers alike, accelerating time to insights while reducing costs.

Get started today!

SQL DI Pro represents a new era of AI-powered analytics for your mission-critical transactional data on the mainframe where advanced AI is accessible, practical, and seamlessly integrated with your existing infrastructure. Whether you're looking to improve customer insights, detect fraud, manage risks, or optimize operations, SQL DI Pro empowers you to unlock the hidden value in your Db2 data.

Ready to discover what your data has been trying to tell you? Visit the IBM SQL Data Insights Pro documentation to learn more.

Guanjun Cai, Senior Content Designer, Db2 for z/OS and Ecosystem Products Content, IBM Software.

Akiko Hoshikawa, Distinguished Engineer, Db2 for z/OS and Al Product Development, IBM Software.

Jason Cu, SQL Data Insights Technical Lead, Db2 for z/OS Development, IBM Software.

#Db2forz/OS
#IBMZ
#z/OS