Skip to main content
PawSQL supports a wide range of mainstream relational databases, Chinese enterprise databases, distributed databases, and big-data SQL engines. Support varies by database and capability. In addition to SQL parsing, PawSQL may provide SQL quality check, query rewrite, index recommendation, execution plan analysis, performance validation, and production SQL governance depending on the target database.

​
Definition

Database support in PawSQL means both the ability to parse a database’s SQL dialect and the depth of database-aware analysis available, which varies by database, version, and deployment mode.

​
Relational Databases

​
Mainstream Databases

  • MySQL
  • PostgreSQL
  • MariaDB
  • Oracle
  • Microsoft SQL Server
  • IBM Db2

​
Chinese Enterprise Databases

  • openGauss
  • GaussDB
  • KingbaseES
  • Dameng
  • MogDB
PawSQL provides database-specific adaptations instead of treating these systems only as MySQL- or PostgreSQL-compatible dialects. Depending on the database, adaptations may cover:
  • SQL syntax and data types
  • Built-in functions
  • DDL syntax
  • Metadata interfaces
  • Indexing mechanisms
  • Query optimizer behavior
  • Execution plan formats

​
Distributed Databases

PawSQL provides dedicated support for distributed database systems where SQL performance depends not only on access paths and indexes, but also on data distribution and cross-node execution. Supported distributed databases include:
  • OceanBase
    • MySQL Mode
    • Oracle Mode
  • TDSQL MySQL
  • TDSQL PostgreSQL
  • TDSQL PostgreSQL Distributed
  • TDSQLx-MySQL
  • GoldenDB
  • PolarDB-X
For distributed databases, PawSQL can analyze database-specific characteristics such as:
  • Distribution keys
  • Sharding keys
  • Sharded tables
  • Replicated or broadcast tables
  • Cross-shard joins
  • Data redistribution
  • Local and global indexes
  • Distributed execution plans
These capabilities can be used in SQL quality check, migration governance, and distributed SQL performance optimization.

​
Big Data

​
Apache Hive

PawSQL supports Apache Hive SQL and provides optimization rules designed specifically for distributed analytical workloads. Supported analysis scenarios include:
  • Partition pruning
  • Bucketed join analysis
  • MapJoin-related optimization
  • COUNT DISTINCT skew analysis
  • GROUP BY skew analysis
  • Window-function skew analysis
  • Global sorting and Top-N optimization
  • UNION skew analysis
  • Distributed SQL best-practice checks
Hive support focuses on SQL quality, execution efficiency, and data-skew risks that are specific to large-scale distributed processing.

​
Compatibility Matrix

The following matrix summarizes the major PawSQL capabilities available across supported databases. Legend
  • ✓ Fully supported
  • ◐ Supported with database- or version-specific limitations
  • — Not applicable or currently unavailable
The matrix represents PawSQL’s overall product capabilities. Exact support may vary by PawSQL release, database version, deployment mode, and whether online metadata access is available.

​
Database Support Is More Than SQL Parsing

For PawSQL, supporting a database does not simply mean that its SQL syntax can be parsed. Database integration may include multiple layers:
The deeper the integration, the more database-aware optimization PawSQL can provide.

​
Online and Offline Modes

PawSQL can work in both online and offline environments.

​
Online Mode

When PawSQL can connect to the target database, it may retrieve:
  • Tables
  • Columns
  • Indexes
  • Views
  • Statistics
  • Distribution metadata
  • Execution plans
Online mode provides the richest context for SQL optimization and performance validation.

​
Offline Mode

When direct database access is restricted, PawSQL can analyze SQL based on schema definitions such as:
  • DDL files
  • Manually imported DDL
  • Schema exported from development or test environments
Offline mode is commonly used for:
  • CI/CD
  • SQL Quality Gates
  • Pre-production review
  • Security-isolated environments
  • Database migration
  • Financial industry environments

​
Version Compatibility

Database capabilities may differ between major and minor versions. Examples include:
  • SQL syntax changes
  • New data types
  • Optimizer behavior
  • Execution plan formats
  • Metadata interfaces
  • Distributed database features
For production deployment, always verify compatibility against the PawSQL release you are using.

​
Compatible Database Variants

Some databases are derived from or compatible with MySQL, PostgreSQL, or openGauss. PawSQL does not automatically assume that compatible databases are identical. Database-specific differences may exist in:
  • Functions
  • DDL
  • Hints
  • Indexes
  • System catalogs
  • Distribution syntax
  • Optimizer behavior
  • Execution plans
Where required, PawSQL maintains dedicated adaptations for individual database products.

​
Feature Availability

Feature availability may depend on:
  • Database type
  • Database version
  • PawSQL edition
  • PawSQL release
  • Online or offline mode
  • Availability of metadata and execution plans
If a feature is marked as partially supported, contact the PawSQL team for the exact compatibility scope of your environment.

​
Need Support for Another Database?

If your database is not listed above, PawSQL may still be able to support it through an existing compatibility layer or a dedicated database adaptation. Typical cases include:
  • MySQL-compatible databases
  • PostgreSQL-compatible databases
  • openGauss-compatible databases
  • Industry-specific databases
  • Enterprise-customized database distributions
Contact the PawSQL team to evaluate compatibility and adaptation requirements.

​
Next steps

After confirming your database and version are supported, follow Choose How to Use PawSQL or the quickstart to start optimizing SQL.