This feature is currently available in early access. To experience it, please submit a ticket or contact your technical advisor.
Are your databases scattered across Alibaba Cloud, on-premises data centers, and third-party cloud providers?
Alibaba Cloud offers a wide range of database products, including PolarDB, ApsaraDB RDS, Tair (Redis OSS-compatible), and ApsaraDB for MongoDB. Each engine has its own monitoring and diagnostic logic tailored to its unique characteristics. When you further layer in SQL Server instances from your on-premises data center and databases from other cloud providers, the result is a chaotic landscape of monitoring platforms, disconnected alert rules, and operations that feel like assembling a puzzle.
Now, there is a solution.
Alibaba Cloud's Database Autonomy Service (DAS) Agent leverages AI capabilities to deliver unified intelligent governance across clouds, engines, and business lines.
Teams no longer rely on "manual inspections"—instead, they use natural language conversations to complete diagnostics, optimizations, and reports.
Teams no longer "step on each other's toes"—they organize by business groups, each with its own dedicated "AI-powered Database Administrator (DBA) assistant."
Since its launch six months ago, DAS Agent has reached millions of daily invocations. Today, let's explore how this paradigm shift in database operations is unfolding.
Enterprise database architectures have long evolved toward hybrid and multi-faceted environments:
• Core transaction systems may be deployed on one public cloud, while overseas operations run on another.
• NoSQL engines like Tair and MongoDB support differentiated business needs such as high concurrency and flexible schemas.
• eanwhile, legacy systems still operate on on-premises data centers or MySQL instances.
• DBA teams also have varied expertise: some excel at MySQL tuning, others specialize in SQL Server high availability, and still others focus on NoSQL architectures.
From monitoring and alerting to performance diagnostics, slow SQL analysis, and capacity planning, each cloud has its own console, each engine has unique diagnostic logic, and each engineer relies on familiar toolchains.
The result?
→ Want to check overall system health? Switch between N different systems.
→ Facing cross-engine issues? Pull together multiple experts for a joint diagnosis.
→ Conducting a full-scale inspection? Write scripts, stitch together data, and manually compile reports.
Operations aren't just a technical challenge—they're a coordination and efficiency problem.
DAS Agent's approach is elegantly simple:
• For Alibaba Cloud's full suite of engines—PolarDB, ApsaraDB RDS, Tair (Redis® OSS-Compatible), ApsaraDB for MongoDB, and more—unified management is available out of the box.
• For on-premises or third-party cloud instances, simply deploy a lightweight, non-intrusive collector from the console (with minimal resource overhead) to integrate them into DAS for unified operations.
Real-World Example: A Leading Fintech Company in Southeast Asia
With 100+ third-party cloud database instances, the company deployed DAS Agent collectors and completed integration in just 1 day. Combined with their 600+ instances on Alibaba Cloud, they achieved unified operations.
The outcome: Following AI-driven recommendations, DBAs and developers implemented targeted solutions. Governance gaps across multi-engine and multi-cloud platforms were drastically reduced. Within one month of deployment, slow SQL queries dropped by 40%.**
This screenshot shows a slow SQL governance platform used by customers via DAS Agent integrated with DMS DIFY and internal systems.
Large enterprises often run multiple business lines in parallel, each with its own database portfolio. How can you ensure isolation while gaining visibility into governance effectiveness?
DAS Agent allows you to create multiple Agents under a single account, with each Agent precisely managing a group of business-related instances. For example:
• Transaction Team: 20 PolarDB instances + 3 Tair instances + 1 MongoDB instance → Configured as "Transaction-Agent"
• Logistics Team: 15 RDS instances + 2 Tair instances → Configured as "Logistics-Agent"
These Agents are isolated in terms of resources, reports, and alerts, yet all leverage the same intelligent platform.
What's more: AI invocation volume = governance activity, and results are quantifiable.
Real-World Scenario
A leading e-commerce platform's technical director can clearly see on the management dashboard: which teams are actively governing, and which databases still carry technical debt. AI isn't a cost center—it's an efficiency amplifier.
This dashboard displays the invocation volume (character count) for each team's Agent. In the details view, you can further review historical diagnostic records and operational reports.
DAS Agent is deeply integrated into key pages of the DAS console. Simply click the AI icon next to any instance to start a professional conversation in natural language. It doesn't just answer—it provides evidence-based, actionable, and comparative optimization recommendations tailored to engine-specific characteristics.
A user is about to deploy a new SQL query to production but is concerned it may exacerbate existing slow SQL issues. Before deploying, they ask the AI to analyze the query in advance and, based on PolarDB MySQL engine characteristics, provide specific, actionable optimization recommendations.
The numbers tell a real story:
• Over the past six months, users have initiated 100,000+ AI diagnostic requests.
• Alibaba Cloud's ticketing support team uses DAS Agent to diagnose database issues with an 83% accuracy rate (industry-standard base model accuracy typically ranges from 40%–60%). The remaining 17% of complex issues are escalated to specialists—maximizing human-AI collaboration efficiency.
| Common Scenario | Query Count |
|---|---|
| Root cause analysis for anomalies like CPU/IO spikes, connection exhaustion, etc. | 69,000+ |
| SQL performance analysis and optimization recommendations | 44,000+ |
| Automated health check reports across multiple instances | 39,000+ |
| Instance resource configuration review and scaling recommendations | 12,000+ |
| NoSQL-specific issues: Tair cache hit ratio, MongoDB slow operations analysis, etc. | 28,000+ |
| Full-stack architecture optimization, high-frequency data caching, etc. | 11,000+ |
| Security risk detection: abnormal logins, SQL injection, and other database security threats | 36,000+ |
These aren't "toy demos"—they're real operational tasks happening daily in production environments.
Chances are, the challenge you're facing has already been solved by DAS Agent tens of thousands of times.
DAS Agent is now fully available for specific users. No deployment required, no modifications needed—simply log in to the DAS console to activate it.
• Create multiple Agents organized by business for team-level isolation
• Unified management of Alibaba Cloud + self-managed + third-party cloud databases
• Out-of-the-box AI diagnostics, inspections, and optimization capabilities
[Infographic] Highlights | Database New Features in February 2026
Alibaba Cloud Indonesia - March 20, 2026
ApsaraDB - June 29, 2020
Alibaba Cloud Native Community - November 6, 2025
Alibaba Cloud Big Data and AI - December 29, 2025
ApsaraDB - February 27, 2026
Alibaba Cloud Big Data and AI - January 21, 2026
Container Compute Service (ACS)
A cloud computing service that provides container compute resources that comply with the container specifications of Kubernetes
Learn More
Container Service for Kubernetes
Alibaba Cloud Container Service for Kubernetes is a fully managed cloud container management service that supports native Kubernetes and integrates with other Alibaba Cloud products.
Learn More
Database for FinTech Solution
Leverage cloud-native database solutions dedicated for FinTech.
Learn More
Oracle Database Migration Solution
Migrate your legacy Oracle databases to Alibaba Cloud to save on long-term costs and take advantage of improved scalability, reliability, robust security, high performance, and cloud-native features.
Learn MoreMore Posts by ApsaraDB
