Why Cloud‑Native Distributed Databases Are the Future of Enterprise Data

Background and Trends

Cloud computing has accelerated the evolution of database systems. From the early 1980s relational databases like Oracle to open‑source MySQL and PostgreSQL, the rapid growth of structured data in the 1990s spurred the development of analytical databases. The 2000‑2010 period marked the rise of big‑data technologies.

Big data emerged due to massive data generation on the Internet and differing processing requirements compared to traditional OLTP workloads. Google’s distributed file system, table storage, and MapReduce papers laid the foundation for today’s big‑data ecosystem. Since 2010, cloud computing has further reshaped data processing, merging traditional relational databases with big‑data ecosystems.

Traditional Von Neumann architectures tightly couple compute and storage, leading to inefficiencies when scaling resources. Cloud computing pools resources via distributed technology, enabling transparent, centralized deployment of applications.

Database Challenges During Peak Loads

During Alibaba’s 2020 Double‑11 event, peak transaction rates reached 580,000 orders per second, translating to several million TPS at the database layer. System load spiked 145‑fold within a second, a level unattainable with legacy technologies alone.

Alibaba Cloud leveraged cloud‑native techniques to meet these demands, achieving the top market share in the Asia‑Pacific cloud database market and entering the global leader quadrant.

Industry Trends

Integration of big data and databases.

Combination of cloud‑native and distributed technologies.

Intelligent automation.

Hardware‑software co‑design for high‑performance data processing.

Security and data immutability.

Key Technologies of Next‑Generation Databases

Core innovations include resource pooling, storage‑compute separation, distributed shared storage, high‑performance I/O stacks (NVM, RDMA), multi‑zone replication, and full‑encryption that allows processing without decryption.

Core Products

PolarDB – Cloud‑Native Relational Database

PolarDB provides three‑node physical replication per data block, offering transparent distributed operation. It is 100 % compatible with MySQL 5.6/5.7/8.0, PostgreSQL 11, and highly compatible with Oracle. Features include:

Storage‑compute separation with minute‑level scaling of compute nodes and storage capacity up to 100 TB.

Intelligent load balancing with transparent read/write separation and customizable access paths.

Distributed shared storage with multi‑TB backup capability.

User‑space I/O stack + NVM + RDMA delivering up to 6× MySQL performance and 1 M QPS.

Redo‑log based replication with millisecond‑level data sync for read‑only instances.

PolarDB‑X – Cloud‑Native Distributed Database

PolarDB‑X extends the cloud‑native architecture with a three‑node distributed design per partition, supporting cross‑AZ deployment and strong consistency via consensus protocols.

New Generation Cloud‑Native Distributed Database System

Combines cloud‑native elasticity and high availability with distributed horizontal scalability, addressing high‑concurrency bottlenecks.

Cloud‑Native Data Warehouse and Data Lake

AnalyticDB implements a cloud‑native data warehouse with pooled storage and compute, enabling elastic, cost‑effective analytics for massive, structured data such as Alibaba’s real‑time transaction logs.

The cloud‑native data lake (DLA) provides a unified interface for heterogeneous data sources on object storage, offering serverless analytics, metadata management, and strong security while keeping data isolated across tenants.

Intelligence, Security, and Ecosystem Tools

Intelligent operations use Kubernetes orchestration for anomaly detection and security diagnostics across multi‑source resources.

Full‑encryption ensures data remains encrypted within the kernel and during big‑data processing, eliminating the need for decryption.

Toolchain includes DTS for data synchronization, DBS for multi‑cloud backup, and DMS for enterprise‑grade data modeling and migration, forming the backbone of Alibaba’s Double‑11 data pipeline.

Best Practices

The pandemic accelerated the convergence of offline and online business, increasing traffic volatility. Cloud‑native databases and data warehouses supported not only Double‑11 but also sectors like online education and gaming, demonstrating resilience and scalability.

The case collection covers seven industries—Internet, new retail, energy, transportation, mobile payment, IoT, and software—showcasing best practices for database selection, rapid deployment, and cost‑effective digital transformation.