Powering Industrial Digitalization | GBase Database AI Integration Solution Showcased at China Industrial Software Conference

On July 31, 2026, the 6th China Industrial Software Conference was held in Chongqing. Under the theme "AI-Integrated Intelligence Empowering Applications, Charting a New Course," the conference brought together over 500 participants from national ministries, municipal government departments, industry associations, research institutes, leading enterprises, investment firms, and media, to explore pathways for integrating industrial software with artificial intelligence.

At the "AI + Industrial Software Industry Integration Development Forum," Mr. Ding Mingfeng delivered a keynote speech titled "Building a Solid Foundation for Industrial Intelligence – GBASE's Full-Stack Database AI Integration Solutions."

I. The Era of Intelligent Industrial Software Demands Advanced Data Management Capabilities

Mr. Ding noted that industrial software has evolved over four decades—from tools replacing manual work to platforms connecting entire ecosystems—and has now entered an intelligence-driven era powered by large language models and knowledge graphs. The greatest bottleneck to AI adoption in industrial scenarios, however, is not that LLMs are insufficiently capable, but rather that operational data, engineering data, and process/compliance systems cannot be unified due to severe data silos.

"Database management systems are the critical infrastructure of China's industrial software value chain," Mr. Ding emphasized. He described databases as the "data heart" of industrial software—indispensable foundational software for smart manufacturing and digital transformation. The current low penetration rate of industrial AI stems primarily from the absence of a robust intelligent data foundation.

II. GBASE's AI Integration Solutions

Addressing this industry challenge, Mr. Ding introduced GBASE's full-stack product capabilities and AI integration strategy—built on the GBase full-stack database portfolio, anchored by three core design principles: "Intelligent Kernel + Intelligent Operations + Intelligent Interaction." The solution integrates with multiple LLMs for model inference and provides standardized connectivity to external AI ecosystems.

At the product capability level, he elaborated on three fundamental transformations databases are undergoing in the AI era:

1. Multi-Model Integration with Native Vector Support

Traditional single-structure data management systems cannot handle the diverse data types found in industrial scenarios. GBASE delivers "one copy of data, multiple access methods" through a unified storage engine, eliminating data silos. The database also embeds native vector engines and vector indexing capabilities, enabling hybrid search and allowing industrial mechanisms to be converted into vector representations for AI agents to retrieve and act upon.

2. Intelligent Interaction

In the past, interacting with databases required specialized SQL expertise. With NL2SQL technology, users can now interact with databases using natural language, dramatically lowering the barrier to entry. Intelligent agents can also call database interfaces on demand, providing query and management capabilities.

3. AI-Powered Autonomous Operations

Traditional database troubleshooting relies heavily on human expertise, with recovery often measured in hours. GBASE leverages AI to deliver self-perceiving, self-diagnosing, self-optimizing, and self-healing "autonomous driving" capabilities, continuously reducing manual operations overhead.

III. From Data Warehouses to AI-Native Industrial Data Foundations

Mr. Ding specifically highlighted that industrial data warehouses are now undergoing lakehouse integration and lakehouse-warehouse convergence. GBASE's lakehouse solution bridges the gap between data lakes and data warehouses, supporting multiple data types at the storage layer—including SQL-structured data, JSON semi-structured data, and object data engines—while providing stream processing, high-performance columnar analytics, and cross-engine data consistency guarantees.

"The goal is to evolve from a 'reporting repository' to a 'contextual layer for AI inference,'" Mr. Ding explained. Data must not only be stored efficiently and queried quickly, but also be directly "understandable" and "invokable" by AI, serving as a reliable foundation for business forecasting and AI agents.

In the latter half of his presentation, Mr. Ding shared GBASE's intelligent solutions in two key areas—industrial knowledge retrieval and predictive maintenance—and showcased multiple national benchmark case studies. These include powering data sharing across China's State Grid dispatch cloud platform, ensuring the stable operation of the national pipeline network's core SCADA system, and covering nearly 100 subway lines across 30 cities. These deployments validate the maturity and reliability of GBase databases in mission-critical, core industry scenarios.

IV. GBASE's Future Direction

Mr. Ding closed his presentation by outlining three strategic directions for GBASE:

• Deeper Built-in Intelligence – Embedding AI capabilities deeper into the database kernel to automatically optimize query execution and resource scheduling.

• Broader Ecosystem Connectivity – Ensuring compatibility with mainstream tools, connecting cloud, edge, and diverse endpoint systems, and opening APIs to support industry integration.

• Stronger Scenario-Specific Deployment – Supporting mission-critical business continuity, enabling real-time analytics and intelligent decision-making, and validating high availability in complex production environments.

From "building a solid foundation for industrial intelligence" to "making data accessible to AI," Mr. Ding's presentation clearly charted GBASE's evolution from a traditional database provider to an intelligent data foundation for the AI era. Going forward, GBASE will continue to focus on product excellence and ecosystem development, ensuring that China's own data infrastructure is robust enough to support the future of industrial intelligence.