This topic describes how to use the Performance Insight function for load monitoring, association analysis, and optimizing performance. This function helps you quickly evaluate the loads of your ApsaraDB for RDS instance and locate performance problems to ensure database stability.
Prerequisites
- Your RDS instance runs one of the following database engine versions:
- MySQL 8.0
- MySQL 5.7
- The kernel version of your RDS instance is 20190915 or later.
Note Log on to the ApsaraDB for RDS console, find the target RDS instance, and navigate to the Basic Information page. Then in the Configuration Information section, check whether the Upgrade Kernel Version button is available. If the button is available, click it to view the kernel version of your RDS instance. If the button is not available, you are already using the latest kernel version. For more information, see Update the minor engine version of an ApsaraDB RDS for MySQL instance.
Overview
The Performance Insight function consists of the following two parts:
- Object Statistics
Object Statistics queries statistics from indexes and the following two tables:
- TABLE_STATISTICS: records rows with read and modified data.
- INDEX_STATISTICS: records rows with data read from indexes.
- Performance Point
Performance Point collects performance details of your RDS instance. Using these details, you can quantify the overheads of SQL statements faster and more accurately. Performance Point measures database performance using the following three dimensions:
- CPU: includes but is not limited to the total time spent executing an SQL statement and the time spent by CPU executing an SQL statement.
- Lock: includes the time occupied by locks such as metadata locks on the server, storage transaction locks, mutual exclusions (mutexes) (in debugging mode only), and readers-writer locks.
- I/O: includes the time taken to perform operations such as reading and writing data files, writing log files, reading binary logs, reading redo logs, and asynchronously reading redo logs.