Database Industry Landscape and Technology Trends | GBase Tech Cloud Talks: Expert Session featuring Jiang Chunyu

At the 2026 GBase Cloud Sharing Conference, Jiang Chunyu, Director of the Big Data and Intelligence Department at the Artificial Intelligence Research Institute of the China Academy of Information and Communications Technology (CAICT), delivered a keynote speech titled “Database Industry Landscape and Technology Trends.” From the perspectives of industry development, technology evolution, and application, he systematically reviewed the current state and trends of both the global and China’s database markets, offering forward-looking insights to the attendees.

Industry Landscape: Expanding Scale and Increasing Clarity in Market Structure

Jiang pointed out that the global database market is entering a phase of high-quality development. According to the Database Development Research Report (2025), China’s database market reached $8.37 billion in 2024 (approximately RMB 59.6 billion), accounting for 7.3% of the global market, and is projected to reach RMB 83.742 billion by 2027.

• From the competitive landscape perspective, the total number of global database vendors has slightly decreased, with around 400 product providers currently. The United States and China lead with 146 and 103 vendors, respectively. China’s database industry has shifted from rapid growth to high-quality development, with an increasingly clear market structure and strong head effects. In terms of product types, non-relational and hybrid databases dominate globally, with key-value databases firmly in first place; while China still mainly focuses on relational databases, but vector databases are gaining increasing attention each year.

• In terms of business models, commercial databases dominate globally, and China has a large share of commercial databases with broad prospects for open-source databases. PostgreSQL ecosystem companies are highly favored by the capital market, with their ease of use and compatibility continuously acknowledged as strong attractions. Multi-cloud management and “AI+” have become focal points for database investment and financing.

• In academic innovation, multi-model data processing has become a key research focus in the database field. Judging from paper distributions at the three top academic conferences (SIGMOD, VLDB, ICDE), research related to non-relational databases has surpassed that of relational databases for consecutive years. Topics such as vector databases, cloud-native databases, graph neural networks, and “AI+database” continue to gain momentum. China's influence at global database academic conferences has steadily grown, with papers from China accounting for an average of over 40% of the total at these three conferences annually, showing an upward trend. Enterprises and universities are increasingly emphasizing the integration of theoretical innovation and practical application.

• From the perspective of standardization, China has made orderly progress in database standardization. The Big Data Technology Standards Promotion Committee of the China Communications Standards Association (CCSA TC601) has released 40 standards since 2015, gradually building a standard system targeting database products, services, and applications. Standards such as the “Database Service Capability Maturity Model” for service capabilities, and the “Database Operation and Maintenance Management Capability Maturity Model” for industry applications, have been adopted by many organizations, effectively guiding the industry towards high-quality development.

Technology Trends: AI and Databases Moving Toward Bidirectional Integration

Jiang Chunyu provided an in-depth analysis of the evolution direction of database technology. At the architectural level, databases are moving from “divide and conquer” to a unified approach—HTAP architecture integrates transaction processing and analytical processing into a single system, breaking down the barriers between traditional OLAP and OLTP. At the same time, cloud-native deployment has become a mainstream trend, achieving efficient resource utilization and elastic scaling through the separation of storage and compute.

Database technology is now fully entering the “AI-native era.” Jiang elaborated on the two-way path of AI and database integration:

AI for DB
AI is Reshaping the Database Kernel

From query optimizers to index management and join ordering, traditional rule-based static optimization is being replaced by learned optimizers. At the same time, AI has greatly enhanced the development and operation efficiency of databases: Text2SQL allows users to perform data queries without mastering complex syntax; intelligent monitoring and alerting, automated inspection and diagnosis, and fault self-healing capabilities are pushing database operations from passive response towards intelligent autonomy.

DB for AI
Databases are Empowering AI Application Implementation

As AI applications move into more complex areas, data requirements have evolved from single types to unified processing of structured, semi-structured, unstructured, and vector data. Vector databases are rapidly maturing, and breakthroughs in “vector+scalar” hybrid retrieval technology provide efficient data retrieval support for large models. Databases are shifting from “passive storage” to “active intelligent collaboration,” with Agent-Native databases emerging as a new trend.

In terms of RAG technology evolution, traditional RAG is limited by simple vector similarity search, making it difficult to handle complex queries and multi-document associative analysis. GraphRAG combines the relational reasoning of graph databases with the semantic matching capability of vector retrieval, significantly enhancing the context understanding ability of large models. Furthermore, through layered memory storage, AI applications are moving from “single conversations” to “long-term memory,” supporting tens of billions of memory scale with over 10x performance improvement.

Industry Applications: From “Usable” to “Truly Effective”

Jiang also shared the application status of databases in key industries. Leading users in finance, telecommunications, and energy have widely deployed distributed systems, with operation and maintenance personnel investment increasing year by year. To address distributed operation challenges, CAICT, in collaboration with several enterprises, developed the “Database Operation and Maintenance Management Capability Maturity Model” standard.

In terms of application trends, databases are moving from centralized to distributed to meet high concurrency and multi-center demands, from single-model to multi-model fusion to handle complex business scenarios, and from static storage to intelligent reasoning to support real-time analytics and autonomous decision-making. In key industries such as finance, telecommunications, energy, and transportation, the application scenarios of domestic databases are continuously expanding, moving from “usable” to “truly effective.”

From HTAP to cloud-native, from AI for DB to DB for AI, database technology is entering a new era full of opportunities. As an important force in China’s database industry, GBase (General Data Technology) will continue to be driven by technological innovation, collaborating with upstream and downstream partners to promote the high-quality development of the database industry.