DAS inspection scoring rules-Database Autonomy Service(DAS)-阿里云帮助中心

Database Autonomy Service (DAS) periodically inspects your database instances and assigns a health score to help you understand their operational status. This topic lists the deduction items and scoring rules.

Usage notes

You must enable the Inspection and Scoring feature before you can view the inspection list. For instructions, see Inspection and scoring.
The Inspection and Scoring feature uses a 100-point scale. The system deducts points based on the instance's runtime status, in accordance with the scoring rules. You can view the final score in the console. For more information, see Inspection and scoring.

Scoring rules

For ApsaraDB RDS for MySQL, self-managed MySQL, and PolarDB for MySQL:

Item	Description	Criterion	Condition	Formula
CPU utilization (`cpuUsage`)	The average daily CPU utilization. For multi-core CPUs, the value is normalized to a single-core equivalent.	major	70% <= `cpuUsage` < 80%	1 + (`cpuUsage` - 0.7) * 20
CPU utilization (`cpuUsage`)		critical	`cpuUsage` >= 80%	min(3 + (`cpuUsage` - 0.8) * 30, 10)
Memory usage (`memUsage`)	The average daily memory usage.	major	80% <= `memUsage` < 90%	1 + (`memUsage` - 0.8) * 20
Memory usage (`memUsage`)	The average daily memory usage.	critical	`memUsage` > 90%	min((`memUsage` - 0.9) * 50, 10)
Storage usage Important This deduction item does not apply to self-managed MySQL instances.	The percentage of used storage out of the total available storage.	Available days	`availableDays` <= 30	15 - `availableDays` / 3
		Large tables Note A table is considered large if its tablespace, including fragments, exceeds 50 GB. Important PolarDB for MySQL does not analyze large tables, so this deduction item does not apply.	`bigTableCount` > 0	min(`bigTableCount`, 15)
Connection usage (`connectionRate`)	The average daily connection count as a percentage of the maximum allowed connections.	major	70% <= `connectionRate` < 80%	1
Connection usage (`connectionRate`)		critical	`connectionRate` > 80%	3
IOPS usage (`iopsUsage`)	The average daily IOPS as a percentage of the maximum allowed IOPS.	major	70% < `iopsUsage` < 90%	3
IOPS usage (`iopsUsage`)		critical	`iopsUsage` > 90%	5
Active sessions (`threadRunning`)	The daily count of active sessions.	major	`threadRunning` > min(2 * `cpuCores` + 8, 64)	3
Active sessions (`threadRunning`)	The daily count of active sessions.	critical	`threadRunning` > min(4 * `cpuCores` + 8, 96)	9
Slow SQL statements (`slowSqlCount`)	The total number of slow SQL statements generated per day. Note Click the value to view the top 5 slow SQL statements by execution count.	minor	0 < `slowSqlCount` < 100	1 + (`slowSqlCount` - 10) / 30
		major	100 <= `slowSqlCount` < 500	4 + (`slowSqlCount` - 100) / 30
		critical	`slowSqlCount` >= 500	min(18 + (`slowSqlCount` - 500) / 30, 30)
SQL security audit	The number of high-risk SQL statements and SQL injection events.	High-risk SQL statements	`riskSqlCount` > 0	min(`riskSqlCount`, 5)
SQL security audit		SQL injection	`sqlInjectionCount` > 0	min(`sqlInjectionCount`, 5)
Deadlock	Whether a deadlock occurred during the day.	minor	A deadlock is detected.	3

For Redis:

Item	Description	Criterion	Condition	Formula
CPU utilization (`cpuUsage`)	The average daily CPU utilization. For multi-core CPUs, the value is normalized to a single-core equivalent.	major	70% <= `cpuUsage` < 80%	1 + (`cpuUsage` - 0.7) * 20
CPU utilization (`cpuUsage`)		critical	`cpuUsage` >= 80%	min(3 + (`cpuUsage` - 0.8) * 30, 10)
Memory usage (`memUsage`)	The average daily memory usage.	major	80% <= `memUsage` < 90%	1 + (`memUsage` - 0.8) * 20
Memory usage (`memUsage`)	The average daily memory usage.	critical	`memUsage` > 90%	min((`memUsage` - 0.9) * 50, 10)
Connection usage (`connectionRate`)	The average daily connection count as a percentage of the maximum allowed connections.	major	70% <= `connectionRate` < 80%	1
Connection usage (`connectionRate`)		critical	`connectionRate` > 80%	3
Traffic usage	The average daily traffic as a percentage of the maximum allowed traffic.	major	80% <= `FlowRate` < 85% (including `inFlow` and `outFlow`)	1
		critical	`FlowRate` > 85%	3
		critical	`FlowRate` > 90%	5
Hot key	Whether hot keys were detected.	minor	Hot keys were detected.	3
Slow SQL statements (`slowSqlCount`)	The total number of slow SQL statements generated per day. Click the value to view the top 5 slow SQL statements by execution count.	minor	0 < `slowSqlCount` < 100	1 + (`slowSqlCount` - 10) / 30
		major	100 <= `slowSqlCount` < 500	4 + (`slowSqlCount` - 100) / 30
		critical	`slowSqlCount` >= 500	min(18 + (`slowSqlCount` - 500) / 30, 30)
Memory fragmentation ratio	The memory fragmentation ratio is calculated as follows: `mem_fragmentation_ratio` = `used_memory_rss` / `used_memory` `used_memory_rss`: The physical memory that the operating system allocates to Redis, which includes memory fragments. `used_memory`: The memory that Redis uses to store data. For more information about Redis memory fragmentation and the memory fragmentation ratio, see Redis memory fragmentation.	minor	`HighMemoryFragment` > 1.5	min(`HighMemoryFragment` * 2, 5)