PolarDB for MySQL offers two editions: Enterprise Edition and Standard Edition. Both share a broad set of capabilities across cluster management, scaling, high availability, security, and monitoring. The key differences are in advanced cluster topologies, AI integration, and certain storage and performance features that require specific configurations.
Enterprise Edition is built for workloads that need global disaster recovery, AI-driven database management, or more than 7 read-only nodes. It supports Multi-master Cluster (Limitless) Edition, global database networks (GDNs), and PolarDB for AI.
Standard Edition covers the majority of production workloads. It supports the YiTian ARM architecture and provides a cost-effective starting point with all core capabilities.
Feature comparison
Cluster management
| Feature | Description | Enterprise Edition | Standard Edition |
|---|
| x86 architecture | Pairs with Intel processors and high-performance network infrastructure to deliver enhanced performance and stability for enterprise workloads. | Supported | Supported |
| YiTian ARM architecture | Uses YiTian 710 chips developed by Alibaba Cloud and 25 Gigabit Ethernet smart network interface cards (NICs) to provide strong computing power. | Not supported | Supported (MySQL 8.0.1 only) |
| Cluster Edition (1 primary + read-only nodes) | Distributes workloads across one primary node and up to 15 (Enterprise) or 7 (Standard) read-only nodes. The primary node handles read and write requests; read-only nodes handle read requests. Uses active-active failover between nodes. | Up to 15 read-only nodes | Up to 7 read-only nodes |
| Multi-master Cluster (Limitless) Edition | Runs multiple primary nodes to support concurrent writes from different compute nodes. Allows dynamic primary node switching within seconds. | Supported (MySQL 8.0 only) | Not supported |
| GDN | A global database network (GDN) links multiple PolarDB clusters across regions within a country. All clusters replicate data and handle read requests. Write requests are forwarded to the primary cluster for geo-disaster recovery. | Supported | Not supported |
| X-Engine Edition | Built on the log-structured merge-tree (LSM tree) architecture with the Zstandard (ZSTD) compression algorithm. Reduces storage usage by up to 70% compared to InnoDB. | Supported (MySQL 8.0 only) | Supported (MySQL 8.0 only) |
| Cluster recycle | Stores released PolarDB clusters. Restore a released cluster to a new cluster, or delete its backup sets. | Supported | Supported |
| Parameter management | Modify cluster and node parameters in the PolarDB console. Export modified parameters as a template and apply it to clusters in the same region. | Supported | Supported |
| Version management | Update PolarProxy, the database engine, or both independently across the three architectural layers of a PolarDB cluster. | Supported | Supported |
Scaling
| Feature | Description | Enterprise Edition | Standard Edition |
|---|
| Add or remove read-only nodes | Manually add read-only nodes with the required specifications or remove nodes you no longer need. | Supported | Supported |
| Specification change | Change cluster specifications online without locking the cluster. Supports compute scale-up/down and storage scale-in/out. | Supported | Supported |
| Serverless | Scales resources automatically based on workload, removing the need for manual capacity planning. | Supported | Supported |
High performance
| Feature | Description | Enterprise Edition | Standard Edition |
|---|
| IMCI | In-Memory Column Index (IMCI) enables hybrid transactional and analytical processing (HTAP) by supporting OLAP queries on the same cluster as OLTP workloads. | Supported (MySQL 8.0 only) | Supported (MySQL 8.0 only) |
| ePQ | Elastic parallel query (ePQ) runs queries in parallel across a single node or across multiple nodes in the cluster with adaptive scheduling. | Supported (MySQL 8.0 only) | Supported (MySQL 8.0.1 only) |
| SQL query optimizations | Includes subquery decorrelation, left join elimination, join condition pushdown, cost-based query transformation, computing pushdown, Partial Result Cache (PTRC), and IN predicate conversion. | Supported (MySQL 8.0 only) | Supported (MySQL 8.0 only) |
| DDL execution optimizations | Includes instant ADD COLUMN, parallel DDL, DDL I/O performance optimization, faster TRUNCATE/DROP TABLE, nonblocking DDL, DDL physical replication optimization, async metadata lock replication, prevention of long-running transactions blocking DDL, preemptible DDL, and DDL execution status visibility. | Supported | Supported |
| High concurrency optimization | Includes Concurrency Control (CCL), inventory hints, statement queue, and hot row optimization for high-concurrency scenarios. | Supported | Supported |
| Transaction system optimization | PolarTrans uses Commit Timestamp Store (CTS) and Remote Direct Memory Access (RDMA) to improve read/write performance in highly concurrent OLTP scenarios, including global consistency (high-performance mode). | Supported | Supported (PSL4 or PSL5 storage class only) |
| Partitioned table | Fully compatible with native MySQL partition syntax. Supports more partition types and combinations with better performance than native MySQL. | Supported | Supported |
Backup and restoration
| Feature | Description | Enterprise Edition | Standard Edition |
|---|
| Backup and restoration | Supports full data backup and incremental log backup. Restore to any point in time using a full backup set and redo logs. | Supported | Supported |
| Flashback query | Retrieve data from clusters, databases, and tables as it existed at a past point in time. | Supported | Supported |
| Database and table restoration | Restore backup data to new databases or tables without overwriting existing data. Specify new names to restore from one database to another. | Supported | Supported |
High availability
| Feature | Description | Enterprise Edition | Standard Edition |
|---|
| Single-zone high availability | The multi-node architecture automatically fails over to a read-only node when the primary node fails. The read-only node becomes the new primary node. | Supported | Supported |
| Multi-zone high availability | Deploy a cluster across multiple zones to withstand data center-level faults. Change the primary zone as needed. | Supported | Supported |
| Failover with hot standby | Enable hot standby on read-only nodes to improve failover speed and preserve transaction state. | Supported | Supported |
High security
| Feature | Description | Enterprise Edition | Standard Edition |
|---|
| Account management | Manage Alibaba Cloud console accounts and database accounts. | Supported | Supported |
| Cluster whitelists | Restrict cluster access to specific IP addresses or Elastic Compute Service (ECS) instances in configured security groups. | Supported | Supported |
| SSL encryption | Encrypt connections at the transport layer using SSL certificates from certificate authorities (CAs) to protect data in transit. | Supported | Supported |
| TDE | Transparent Data Encryption (TDE) performs real-time I/O encryption and decryption on data files. Encrypts data before writing to disk; decrypts when reading to memory. Does not increase file size and requires no application changes. | Supported | Supported |
| SQL firewall | PolarProxy identifies and blocks SQL statements based on configurable blacklist and whitelist rules. | Supported | Supported |
| Dynamic data masking | Mask sensitive data in real time before exposing it to third parties for reporting, analysis, or development and test activities. | Supported | Supported |
| PolarDB Always-confidential | Implements end-to-end encryption by encrypting data on the client side before it reaches the database. Plaintext data is never visible to the database server. | Supported | Supported |
Connection management
| Feature | Description | Enterprise Edition | Standard Edition |
|---|
| Connection pool | Reduces connection overhead with session-level and transaction-level connection pools. Choose the pool type based on your workload. | Supported | Supported |
| Persistent connection | Maintains stable connections through O&M activities (specification upgrades, switchovers, minor version updates) and server anomalies. | Supported | Supported |
| Consistency levels | Three levels to balance consistency and performance: eventual consistency, session consistency, and global consistency. | Supported | Supported |
| Global consistency (high-performance mode) | PolarTrans uses CTS and RDMA at the kernel level to enforce strong consistency for all read requests forwarded to any read-only node. | Supported | Supported (PSL4 or PSL5 storage class only) |
Cost-effectiveness
| Feature | Description | Enterprise Edition | Standard Edition |
|---|
| Cold data archiving | Archive cold data to Object Storage Service (OSS) to reduce storage costs. | Supported (MySQL 8.0 only) | Supported (MySQL 8.0 only) |
| Data compression | X-Engine Edition, developed by Alibaba Cloud, uses LSM tree architecture to compress archive data and reduce storage costs. | Supported (MySQL 8.0 only) | Supported (MySQL 8.0 only) |
| OSS foreign table | Query cold data archived to OSS directly from PolarDB without moving data back into the cluster. | Supported (MySQL 8.0 only) | Supported (MySQL 8.0 only) |
Monitoring and optimization
All monitoring and optimization features integrate with Database Autonomy Service (DAS).
| Feature | Description | Enterprise Edition | Standard Edition |
|---|
| Diagnosis | Provides autonomy center, session management, real-time monitoring, storage analysis, deadlock analysis, diagnostic reports, and Performance Insight. | Supported | Supported |
| Autonomy center | DAS automatically performs root cause analysis and applies optimizations when exceptions occur, based on your authorization. | Supported | Supported |
| Session management | View session statistics for a cluster. | Supported | Supported |
| Real-time monitoring | Track QPS, TPS, and network traffic for a cluster in real time. | Supported | Supported |
| Storage analysis | View storage overview including available days, tablespace usage, fragmentation percentage, and exception analysis. | Supported | Supported |
| Deadlock analysis | Analyze the most recent deadlock on a database and inspect the analysis details. | Supported | Supported |
| Diagnostic reports | Create and view diagnostic reports with customizable conditions. | Supported | Supported |
| Performance Insight | Aggregate SQL statistics, evaluate cluster workloads, and identify the root causes of performance issues. | Supported | Supported |
| Performance monitoring | Monitor performance metrics at second-level intervals and identify faults from monitoring data. | Supported | Supported |
| Slow SQL query | View slow SQL trends and statistics. Get diagnostic results and optimization suggestions. | Supported | Supported |
| SQL Explorer and Audit | Full-request logging with search, SQL Explorer, security audit, and traffic playback and stress testing. Helps troubleshoot performance issues and identify high-risk sources. | Supported | Supported |
DB for AI
| Feature | Description | Enterprise Edition | Standard Edition |
|---|
| PolarDB for AI | Integrates MLOps and built-in models to deliver a one-stop database service that combines data, features, and models in a single system. | Supported (MySQL 8.0 only) | Not supported |
Data migration and synchronization
| Feature | Description | Enterprise Edition | Standard Edition |
|---|
| Migration from ApsaraDB RDS | Migrate data from an ApsaraDB RDS instance to a PolarDB cluster without changing endpoints. | Supported | Supported |
| Migration from self-managed MySQL | Migrate data from a self-managed MySQL database to a PolarDB cluster. | Supported | Supported |