Help Center/ Relational Database Service_RDS for PostgreSQL/ User Guide/ Problem Diagnosis and SQL Analysis/ Function Overview
Updated on 2026-04-20 GMT+08:00
View PDF
Share

Function Overview

DBA Assistant provides visualized database O&M and intelligent diagnosis for developers and database administrators (DBAs), making database O&M easy and efficient. By analyzing alarms, resources, and performance metrics, it helps users quickly locate faults and keep track of instance status.

To use DBA Assistant, IAM users must have the RDS FullAccess, DAS FullAccess (or DAS Administrator), and CES FullAccess permissions. For details, see Using IAM Roles or Policies to Grant Access to RDS.

In certain regions, to use DBA Assistant, submit a service ticket to request required permissions.

Scenarios

  • In emergency cases, you can manually terminate slow sessions to recover your instance, improving database availability.
  • If your DB instance is unstable due to a large number of concurrent SQL requests from new services, you can set SQL throttling rules for SQL statements to limit concurrent SQL statements and ensure instance stability.
  • If your instance storage is full, you can learn about the storage usage and disk space distribution on the Storage Analysis page. You can enable storage autoscaling. When the available storage of your instance drops to the threshold, autoscaling is triggered. For details, see Configuring Storage Autoscaling.

Functions

Table 1 lists the functions supported by DBA Assistant.

Table 1 Function description

Function

Description

Reference

Overview

Shows the status of your instance, including alarms, resource usages, and key performance metrics. DBA Assistant diagnoses instance health using operational data analytics and intelligent algorithms, and provides you with solutions and suggestions for handling detected exceptions.

Viewing the Overall Status of a DB Instance

Performance

Displays key metrics of your instance and provides metric comparison between different days. You can keep track of metric changes and detect exceptions in a timely manner.

Viewing Performance Metrics of a DB Instance

Sessions

You can query session snapshots of your instance while sorting, filtering, and displaying the snapshots as needed. You can also view session statistics by user, source, and database. Sessions can be killed for urgent instance recovery to ensure database availability.

Managing Real-Time Sessions

Historical Session Analysis

You can query historical sessions, historical wait events, and historical top SQL statements of your DB instance.

Managing Historical Sessions

Storage Analysis

Data and log storage space and its changes significantly impact database performance. RDS for PostgreSQL shows you the distribution and changes of used storage on the Storage Analysis page. It also provides storage autoscaling, intelligent tablespace diagnosis, and analysis of top 50 largest databases and tables.

Managing Storage Capacity

Slow Query Logs

Displays slow queries within a specified time period. You can view top slow query logs by user or client IP address, sort statistics, and identify sources of slow SQL statements.

Viewing Slow Query Logs of a DB Instance

SQL Insights

After Collect All Query Logs is enabled, you can filter, search for, and analyze SQL statements in multiple dimensions and gain a comprehensive insight into all SQL statements. Top SQL helps you locate exceptions.

Creating a SQL Insights Task

SQL Throttling

SQL Throttling restricts the execution of SQL statements based on specified rules when there are SQL statements that cannot be optimized timely or a resource (for example, vCPU) bottleneck occurs.

Creating a SQL Throttling Rule
Parent Topic: Problem Diagnosis and SQL Analysis