How China's DBA Landscape Is Evolving with Domestic Databases and AI

Introduction: A Silent Revolution

By 2025, China’s database market has moved beyond the slogan of “domestic substitution.” Accelerated core‑system migration and deepening Xinchuang policies have pushed domestic databases from merely usable to truly user‑friendly, with penetration in financial core systems exceeding 30% in 2024, up from 5% in 2020.

The transformation is driven by countless DBAs working day and night, covering products, ecosystem tools, services, security, and community, offering diverse career paths and a comprehensive upgrade of DBA capabilities.

1. Database Technology Enters Deep Water: DBA Role Shifts from "Firefighter" to "Architect"

1. Complex Scenarios Demand Skill Upgrades

Domestic databases such as TiDB, OceanBase, PolarDB, GaussDB, and TDsql break foreign monopolies but introduce diverse stacks. DBAs must master multi‑replica consistency (Raft/Paxos), compute‑storage separation, HTAP, and multimodel architectures. For example, Zhihu’s team solved a 34.4‑billion‑record migration on OceanBase by designing index‑based data extraction, requiring deep knowledge of migration techniques and underlying logic.

2. Toolchain Innovation: From Manual to Intelligent

TiDB’s TiUP and OceanBase’s OCP illustrate a three‑step evolution: standardization → automation → intelligence. Future AI‑driven autonomous operations (e.g., TiDB’s smart tuning, OceanBase’s self‑healing) will enable DBAs to move from manual monitoring to data‑driven decision making, effectively achieving “unattended” database management.

3. Paradigm Shift in Performance Tuning

Traditional Oracle tuning (index optimization, execution‑plan analysis) remains relevant, but domestic databases require new approaches. PolarDB’s three‑layer decoupled architecture, AI‑driven autonomous operations, and hardware‑software co‑innovation overcome classic master‑slave bottlenecks. DBAs must now handle hardware tuning, cross‑node caching, and high‑concurrency transaction processing to meet strict RTO and RPO targets in finance.

2. Co‑building the Database Ecosystem: From Single‑Operation to Full‑Stack Collaboration

1. Cross‑Vendor Collaboration Skills

Open ecosystems, such as Huawei storage partnering with Wanli Database, demand DBAs understand both database engines and underlying storage (e.g., Huawei OceanStor) to achieve multi‑master performance breakthroughs, positioning DBAs as “bridge talents” between vendors, storage providers, and developers.

2. Value of Open‑Source Community Participation

Open‑source drives domestic database growth. TiDB’s community exemplifies a “product‑user‑contributor” flywheel, creating a symbiotic ecosystem of technology, talent, and business. Active contributions—code commits, documentation, case studies—accelerate iteration. The author, a core member of TUG and OUG, contributed to TiDB and the ob‑operator automation, enhancing personal influence and corporate experience.

3. Multi‑Tenant and Resource Isolation Practices

OceanBase’s fine‑grained resource isolation combines cgroup and distributed sharding to guarantee CPU, memory, IOPS, and storage exclusivity. Dynamic elastic management via Unit resources enables second‑level scaling, balancing reservation and sharing. For HTAP mixed workloads, dual isolation (physical replica + logical bandwidth limits) ensures low‑latency OLTP and stable AP queries, pushing DBAs toward the role of “business architecture advisors.”

3. DBA Career Leap: From Executor to Decision‑Maker

1. Core Role in Large‑Scale Migration

In massive domestic migration projects, DBAs become central architects. The Zhihu case of migrating >20 TB daily and >1 M QPS to TiDB demonstrates a repeatable methodology, granting DBAs authority over storage engine selection, transaction model design, and technical standards.

2. Cost Optimization and Value Visibility

DBAs must build data‑driven value proof systems. Zhihu’s migration to OceanBase using OCP achieved minute‑level deployment, full‑stack monitoring, and SQL diagnostics, boosting operational efficiency by 60%. TiDB‑operator on Kubernetes delivered elastic scaling, cutting hardware costs by 35%, illustrating concrete “cost‑saving and efficiency‑gain” outcomes.

3. Technical Evangelism and Knowledge Consolidation

Through active participation in TiDB, OceanBase, PolarDB, Pika, MongoDB communities and publishing over 40 technical articles via the “Xiaolei Talks DB” channel, the author built a comprehensive knowledge base covering distributed selection, tuning, multi‑cloud, high‑availability, and FinOps, encouraging DBAs to grow personal IP via contributions and public speaking.

4. Future Outlook: Fusion of AI and Cloud‑Native Technologies

1. AI‑Driven Autonomous Operations

AI for DB reshapes DBA work: PolarDB’s smart tuning and GaussDB’s AI diagnostic tools predict hardware failures, requiring DBAs to grasp basic AI model principles and collaborate with algorithm teams.

2. Cloud‑Native and Serverless Adoption

Kubernetes drives containerized databases; Serverless further reduces operational burden. Zhihu’s DB‑on‑Kubernetes implementation achieves second‑level scaling and cross‑cloud disaster recovery, demanding DBAs master Kubernetes architecture, troubleshooting, and operator tools for various databases.

3. Global Competition and New Battlegrounds

Domestic databases expanding overseas (e.g., TiDB’s international launch) create opportunities for DBAs to handle multi‑region compliance (GDPR), cross‑time‑zone collaboration, and hybrid‑cloud architectures.

Conclusion: Fully Embrace the Domestic Database Wave

The deepening of domestic database substitution is a dual revolution of technology and profession. DBAs transition from fire‑fighters to ecosystem builders, business value translators, and industry leaders. Focusing on five core directions—breaking traditional ops mindsets, mastering HTAP and multimodel selection, advancing cloud‑native skills, contributing to open‑source, and quantifying FinOps impact—will ensure DBAs thrive in the AI‑driven, cloud‑native future.