Global Database Industry Trends and Opportunities for Chinese Vendors

Four trends in the global database industry:

1) Relational databases still dominate while NoSQL databases gradually rise.

2) Cloud providers are increasingly developing their own databases.

3) Open‑source databases are becoming more important.

4) Multi‑model databases are emerging to meet complex application scenarios.

Domestic database vendors benefit from three opportunities:

1) The "Xinchuang" strategy promotes a domestic ecosystem.

2) The big‑data era creates demand for unstructured data processing and high‑parallelism, offering a chance for Chinese vendors to accelerate.

3) Major Chinese cloud providers entering the market reshape competition and break the historic monopoly of giants like Oracle.

Databases are a core layer of infrastructure, alongside operating systems and middleware; virtually all applications rely on databases for data storage, management, and processing, making the choice of database critical for performance, scalability, flexibility, and reliability.

As data‑processing needs evolve—from early consistency‑focused workloads to the rise of the Internet, cloud computing, and AI—databases have shifted from centralized to distributed architectures, and the demand for low‑cost solutions has spurred the growth of open‑source databases. Many cloud vendors now offer their own database products, forming a new competitive force.

Databases can be classified by data structure (relational vs. non‑relational) and by function (operational vs. analytical).

Different database types emphasize different strengths, advantages, and suitable domains; no single product fits all scenarios.

Operational/transactional databases handle two kinds of data: persistent data tied to applications (e.g., ERP, CRM) that require stable storage, and temporary data generated during runtime that often uses NoSQL solutions due to lower consistency requirements.

Analytical databases are not directly linked to a specific application; they are used in data warehouses, business intelligence, and data‑science contexts, typically containing data extracted from operational databases for large‑scale analysis.

Relational databases are built on the relational model (one‑to‑one, one‑to‑many, many‑to‑many) and use SQL. Major examples include Oracle, Microsoft SQL Server, MySQL, PostgreSQL, DB2, SQLite, MariaDB, SAP, and others.

According to DB‑Engines Ranking, relational databases dominate the top‑10 popularity list, with Oracle, MySQL, and Microsoft SQL Server occupying the top three positions.

Relational databases excel in data consistency (transactional integrity) and have low update overhead due to standardized schemas, but they struggle with horizontal scalability, flexibility, and handling large volumes of unstructured data or high‑concurrency workloads.

NoSQL databases address unstructured data, high concurrency, and flexible scaling. They support non‑SQL query languages and include document, graph, key‑value, time‑series, and wide‑column stores.

NoSQL advantages are ease of use, horizontal scalability, and strong handling of unstructured data, often providing API‑level access that abstracts underlying structures; however, they may suffer from data inconsistency and are less mature than relational systems.

Commercial database market share remains concentrated among a few giants: Oracle, Microsoft, and IBM together hold 79% of relational‑database revenue, while Microsoft, IBM, and InterSystems dominate the non‑relational commercial segment with a combined 76% share.

Domestic database technology dates back to the 1970s, with early self‑development efforts; the rise of the Internet popularized open‑source databases like MySQL, and the subsequent growth of cloud computing spurred Chinese vendors to develop their own cloud‑native databases, gradually reducing the market share of overseas giants.

China’s “old four” domestic database vendors—Dameng, Renmin University Jincang, Nanda General, and Shenzhou General—are now joined by new giants such as Alibaba OceanBase and Huawei GaussDB. They face pressure from international incumbents but benefit from opportunities created by big‑data, high‑concurrency, and AI‑driven demands, as well as the national "Xinchuang" strategy.

Dameng, founded in 2000 under the China Electronics Corporation, offers database products, big‑data platforms, and open cloud platforms, with capabilities comparable to Oracle.

Renmin University Jincang, established in 1999, provides solutions for e‑government, defense, finance, smart cities, and enterprise digitalization.

Nanda General, founded in 2004, delivers the GBase family of databases and has entered Gartner’s Magic Quadrant for analytical data management.

Shenzhou General, part of China Aerospace Science and Technology Corporation, supplies relational, KStore, xCluster, and BI suites, serving sectors such as government, telecom, energy, transportation, cybersecurity, defense, and aerospace.

In summary, the rise of big data and AI increases demand for high‑concurrency, flexible, and unstructured‑data handling; cloud providers’ self‑developed databases, the growth of open‑source solutions, and the weakening monopoly of traditional relational giants create space and opportunities for Chinese database vendors.

Source: "国产数据库：机遇和挑战深度分析"

Download link: 国产基础软硬件：开源、迁移、上云，关键在生态

Related downloads include frameworks for GPU, CPU, ARM, and Chinese CPUs, as well as industry deep‑reports.

Promotional note: The full collection of architecture‑related technical materials (35 books) is available for purchase at ¥188 (originally ¥290), with free updates for buyers.