Storage Analysis

Checking the Storage Overview

1. Log in to the DAS console.
2. Click in the upper left corner and select a region and project.
3. In the navigation pane, choose Intelligent O&M > Instance List.

   Alternatively, on the Overview page, click Go to Intelligent O&M.

4. In the upper right corner of the Instance List page, search for instances by engine type, instance name, or instance IP.
5. Locate the target instance and click Details. The Dashboard page is displayed.
6. Click the Storage Analysis tab and check information in the Overview area. If the capacity is insufficient, scale up the storage or enable autoscaling.
   Figure 1 Overview
   

   If the average daily increase in last week is 0 GB, the estimated available days of storage are unlimited and are not displayed.

   If the total number of files in your storage (including data space, binlog space, slow query log space, relay log space, audit log space, temporary space, and other space) exceeds 10,000, DAS will not collect information about the files or display space distribution and usages over time on the console. This prevents performance slowdowns caused by collecting statistics on too many files. If this happens, submit a service ticket.

   • Data: Disk space occupied by user data (including temporary table files and ib_logfile files generated by the database)
   • Binlogs: Disk space occupied by binlogs
   • Slow query logs: Disk space occupied by slow query logs
   • Relay logs: Disk space occupied by relay logs
   • Audit logs: Disk space occupied by audit logs
   • Temporary space: Disk space occupied by temporary files
   • Other: Disk space occupied by files such as ibdata1, ib_buffer_pool, ib_doublewrite, and error.log, plus the reserved space (about 5% of the disk space) for OS

Viewing the Disk Space Change Trend

You can check the storage usage over time of your instance.

Figure 2 Disk space distribution

Tablespaces

DAS can diagnose tables with abnormal tablespace growth, tables without primary keys, and tables without indexes. To use this function, subscribe to Intelligent O&M first.

1. Click the Storage Analysis tab to view abnormal tables.
   Figure 3 Tablespaces
   

2. Click Subscribe. In the displayed dialog box, you can learn about enhanced features and billing rules.
   Figure 4 Subscribing to Intelligent O&M
   

3. After subscribing to Intelligent O&M, view the table diagnosis results of your instance.
   Figure 5 Table diagnosis results
   

4. Click next to Auto Diagnosis. In the displayed dialog box, configure the daily tablespace increase limit and click OK.
   Figure 6 Configuring a daily tablespace increase limit
   

Top Databases and Tables by Physical File Size

You can view the top 50 databases and tables by physical file size and identify the databases and tables with high usage based on storage space distribution.

Figure 7 Top 50 databases and tables

Click View Chart to view data volume changes in the last 7 days, last 30 days, or a custom time period (spanning no more than 30 days).

Figure 8 Viewing the trend

• DAS collects data at about 04:00 every day.
• Physical file sizes are precisely recorded, but other fields' values are estimated. If there is a large gap between a file size and another field, run ANALYZE TABLE on the table.
• A database or table whose name contains reserved special characters, including slashes (/) and #p#p, is not counted.
• Currently, only RDS for MySQL 5.7 and 8.0 are supported.
• If your instance's memory usage is greater than 85% or it has more than 50,000 tables, DAS will not collect related information. This prevents performance slowdowns caused by collecting statistics on too many databases and tables.
• Some statistics may be missing because of database or table data fluctuations.

Table Defragmentation

Among the top 50 tables, you can defragment non-system tables. Single-node instances support table defragmentation by default. To use this function for primary/standby instances, submit a service ticket.

• Do not perform defragmentation when the instance is executing other tasks. Otherwise, these tasks may fail.
• Defragmentation tasks for tables are mutually exclusive and must be executed in a queue to reduce impact on instance performance.
• A table is defragmented by running the OPTIMIZE statement on it. During this process, the table is temporarily locked.
• Defragmenting a table often requires significant time and resources. You need to reserve storage space that is at least 1.5 times the size of this table.
• Canceling an ongoing defragmentation task will damage MyISAM tables and make them unavailable. Exercise caution when performing this operation.
• After a table defragmentation is complete, the system will update fragmentation statistics for the table.
• When you configure defragmentation for a primary/standby or cluster instance, if Standby node only is not selected, the system will defragment tables on all nodes in the instance. If Standby node only is selected, only tables on standby nodes will be defragmented.

In the Top 50 Tables list, find the target table and click Defragment in the Operation column. In the displayed dialog box, select an option for Executed and click OK. Then, a confirmation prompt appears. Click OK.

Figure 9 Selecting the Now option

Figure 10 Selecting the Maintenance window option

For more operations, see Viewing a Task.