Metrics - ApsaraDB for Redis - Alibaba Cloud Documentation Center

ApsaraDB for Redis provides a variety of metrics to help you monitor the status of your ApsaraDB for Redis instances in real time. This topic describes these metrics and their limits.

View monitoring data

For more information, see View monitoring data.

Monitoring groups for data nodes

Metrics in basic monitoring groups and commands-related monitoring groups are supported for data nodes of an ApsaraDB for Redis instance. For cluster instances and read/write splitting instances, specific metrics are displayed in aggregate views. For example, if the CPU utilization data of all data nodes is aggregated, the average CPU utilization of data nodes in the current instance is displayed.

Note For read/write splitting instances, the metric data of read replicas is not aggregated.

Table 1. Basic monitoring groups
Monitoring group	New metric name	Old metric name	Support for aggregate views	Unit	Description and limit
CPU Utilization	CPU Utilization	CpuUsage	✔️	%	N/A
Used Memory, Memory Usage	Memory Usage	memoryUsage	✔️	%	This metric is available only for instances that run Redis 4.0 or later. For more information about how to upgrade the major version of an ApsaraDB for Redis instance, see Upgrade the major version.
	Used Memory	UsedMemory	✔️	Bytes	The amount of used memory, which includes the memory consumed by data and cache.
	TairPENAStandardPmemUsage	N/A	✔️	%	This metric is available only for persistent memory-optimized instances of ApsaraDB for Redis Enhanced Edition (Tair). For more information about persistent memory-optimized instances, see Persistent memory-optimized instances.
	Persistent Memory Usage	N/A	✔️	Bytes
	Memory Usage of Lua Scripts	UsedMemoryLua	❌	Bytes	N/A
Disk Usage	Disk Usage	N/A	❌	%	This metric is available only for ESSD-based instances of ApsaraDB for Redis Enhanced Edition (Tair). For more information about ESSD-based instances, see ESSD-based instances.
Requests	Total QPS	TotalQps	✔️	Count/s	The total number of requests per second, which includes read and write requests.
	Read QPS	GetQps	✔️	Count/s	The number of read requests per second and the number of write requests per second. These metrics are available only for instances that run Redis 4.0 or later.
	Write QPS	PutQps	✔️	Count/s
	Other QPS	None	✔️	Count/s	The number of non-read/write requests, such as PING, INFO, SELECT, and AUTH requests. This metric is available only for instances that run Redis 4.0 or later. Note For instances that contain proxy nodes, the Other QPS metric collects only the number of INFO requests.
Connections, Connection Usage	Connection Usage	ConnectionUsage	✔️	%	The connection usage is calculated by using the following formula: Connection usage = Number of used connections/Number of supported connections. For more information about connections that are supported for an ApsaraDB for Redis instance, see Overview. Note If clients connect to a cluster instance in direct connection mode, you must pay attention to this metric. If clients connect to a cluster instance or read/write splitting instance by using a proxy node, you can ignore this metric and instead pay attention to proxy node metrics. For more information, see Monitoring groups for proxy nodes.
Connections, Connection Usage	Used Connections	ConnCount (UsedConnection)	✔️	N/A
Outbound/Inbound Traffic Rate	Inbound Traffic Rate	IntranetIn	✔️	Kbit/s	N/A
Outbound/Inbound Traffic Rate	Outbound Traffic Rate	IntranetOut	✔️	Kbit/s	N/A
Outbound/Inbound Traffic Usage	Inbound Traffic Usage	IntranetInRatio	✔️	%	N/A
Outbound/Inbound Traffic Usage	Outbound Traffic Usage	IntranetOutRatio	✔️	%	N/A
Latency	Average Latency	AvgRt	✔️	μs	The average response time of all commands. This is the average time that is measured from the time when a data node receives a command request to the time when the data node returns a response.
Key Statistics	Total Keys	Keys	✔️	N/A	The total number of primary keys that are stored in the instance.
	Total Keys with TTLs	Expires	✔️	N/A	The instantaneous value of the total number of keys when the data is collected.
	Total Purged Keys	ExpiredKeys	✔️	N/A	N/A
	Total Evicted Keys	EvictedKeys	✔️	N/A
	Purged Keys Per Second	ExpiredKeysPerSecond	✔️	Count/s
	Evicted Keys Per Second	EvictedKeysPerSecond	✔️	Count/s
Hit Ratio	Hit Ratio	hit_rate	✔️	%	The hit ratio of keys. The value is calculated by using the following formula: Hit ratio = Total key hits/(Total key hits + Total key misses). Note If you use an ApsaraDB for Redis instance whose major version is Redis 2.8, this metric is not displayed. You must update the minor version of the instance. For more information, see Update the minor version.
Hits and Misses	Hits Per Second	hit	✔️	N/A	N/A
Hits and Misses	Misses Per Second	miss (Miss Count)	✔️	N/A	N/A
Disk Information	Total Disk Usage	ins_size	✔️	MB	This metric is available only for hybrid storage instances of ApsaraDB for Redis Enhanced Edition (Tair). For more information about hybrid storage instances, see Hybrid-storage instances (phased out).
	Disk Usage of Data Files	DataSize	✔️	MB	The amount of disk space that is occupied by data files, which includes append-only files (AOF) and Redis Database (RDB) files. Note This metric is available only for DRAM-based instances of ApsaraDB for Redis Enhanced Edition (Tair). For more information about DRAM-based instances, see DRAM-based instances.
	Disk Usage of Log Files	LogSize	✔️	MB	The amount of disk space that is occupied by log files, which includes the operational logs of the data node. Note This metric is available only for DRAM-based instances of ApsaraDB for Redis Enhanced Edition (Tair). For more information about DRAM-based instances, see DRAM-based instances.
Statistics of Keys on Disk and in Memory	Keys on Disk	swapped_key	✔️	N/A	These metrics are available only for hybrid storage instances of ApsaraDB for Redis Enhanced Edition (Tair). For more information about hybrid storage instances, see Hybrid-storage instances (phased out).
Statistics of Keys on Disk and in Memory	Keys in Memory	inmem_keys	✔️	N/A

Table 2. Commands-related monitoring groups
Monitoring group	Support for aggregate views	Unit	Description and limit
Key Monitoring Group	❌	Count/s	The monitoring information about key-value related commands, such as the number of times DEL and EXITS are executed.
String Monitoring Group	❌	Count/s	The monitoring information about string commands, such as the number of times APPEND and MGET are executed.
Hash Monitoring Group	❌	Count/s	The monitoring information about hash-related commands, such as the number of times HGET and HDEL are executed.
List Monitoring Group	❌	Count/s	The monitoring information about list commands, such as the number of times BLPOP and BRPOP are executed.
Set Monitoring Group	❌	Count/s	The monitoring information about set commands, such as the number of times SADD and SCARD are executed.
Zset Monitoring Group	❌	Count/s	The monitoring information about zset commands, such as the number of times ZADD and ZCARD are executed.
HyperLog Monitoring Group	❌	Count/s	The monitoring information about HyperLogLog commands, such as the number of times PFADD and PFCOUNT are executed.
Pub/Sub Monitoring Group	❌	Count/s	The monitoring information about publication and subscription commands, such as the number of times PUBLISH and SUBSCRIBE are executed.
Transaction Monitoring Group	❌	Count/s	The monitoring information about transaction commands, such as the number of times WATCH, MULTI, and EXEC are executed.
Lua Script Monitoring Group	❌	Count/s	The monitoring information about Lua script commands, such as the number of times EVAL and SCRIPT are executed.
Tairdoc Monitoring Group	❌	Count/s	The monitoring information about TairDoc commands, such as the number of times JSON.SET and JSON.GET are executed. Note This metric is available only for DRAM-based instances of ApsaraDB for Redis Enhanced Edition (Tair). For more information about DRAM-based instances, see DRAM-based instances.
TairHash Monitoring Group	❌	Count/s	The monitoring information about exHash commands, such as the number of times EXHSET and EXHMSET are executed. Note This metric is available only for DRAM-based instances and persistent memory-optimized instances of ApsaraDB for Redis Enhanced Edition (Tair). For more information about DRAM-based instances, see DRAM-based instances. For more information about persistent memory-optimized instances, see Persistent memory-optimized instances.
TairString Monitoring Group	❌	Count/s	The monitoring information about exString commands, such as the number of times EXSET and EXGET are executed. Note This metric is available only for DRAM-based instances and persistent memory-optimized instances of ApsaraDB for Redis Enhanced Edition (Tair). For more information about DRAM-based instances, see DRAM-based instances. For more information about persistent memory-optimized instances, see Persistent memory-optimized instances.
TairGis Monitoring Group	❌	Count/s	The monitoring information about TairGIS commands, such as the number of times GIS.ADD, GIS.GET, and GIS.DEL are executed. Note This metric is available only for DRAM-based instances of ApsaraDB for Redis Enhanced Edition (Tair). For more information about DRAM-based instances, see DRAM-based instances.
TairBloom Monitoring Group	❌	Count/s	The monitoring information about TairBloom commands, such as the number of times BF.RESERVE and BF.ADD are executed. Note This metric is available only for DRAM-based instances of ApsaraDB for Redis Enhanced Edition (Tair). For more information about DRAM-based instances, see DRAM-based instances.

Monitoring groups for proxy nodes

Metrics for proxy nodes of cluster and read/write splitting instances are collected. Some metrics are displayed in aggregate views. For example, if the CPU utilization data of all proxy nodes is aggregated, the average CPU utilization of proxy nodes in the current instance is displayed.


Monitoring group	New metric name	Old metric name	Support for aggregate views	Unit	Description and limit
CPU	CPU Utilization	CpuUsage	✔️	%	N/A
Requests	Total QPS	TotalQps	✔️	Count/s	The total number of requests per second, which includes read and write requests.
	Read QPS	GetQps	✔️	Count/s	The number of read requests per second and the number of write requests per second. Note To collect this metric, you must update the minor version of the ApsaraDB for Redis instance to the latest version. For more information, see Update the minor version.
	Write QPS	PutQps	✔️	Count/s
Connections	Connection Usage	ConnectionUsage	✔️	%	The connection usage is calculated by using the following formula: Connection usage = Number of used connections/Number of supported connections. For more information about connections that are supported for an ApsaraDB for Redis instance, see Overview. Note If clients connect to a cluster instance or read/write splitting instance by using a proxy node, you must pay attention to this metric. If clients connect to a cluster instance of ApsaraDB for Redis in direct connection mode, you can ignore this metric and instead pay attention to data nodes metrics. For more information, see Monitoring groups for data nodes.
Connections	Used Connections	UsedConnection	✔️	N/A
Outbound/Inbound Traffic	Inbound Traffic Rate	IntranetIn	✔️	Kbit/s	N/A
Outbound/Inbound Traffic	Outbound Traffic Rate	IntranetOut	✔️	Kbit/s	N/A
Size of Requests and Responses	Average Bytes Per Request	AvgRequestSize	❌	Bytes	N/A
	Average Bytes Per Response	AvgResponseSize	❌	Bytes
	Maximum Bytes Per Request	MaxRequestSize	❌	Bytes
	Maximum Bytes Per Response	MaxResponseSize	❌	Bytes
Network latency	Average Latency	AvgRt	✔️	μs	The average response time of all commands. This is the average time that is measured from the time when the proxy node receives a command to the time when the proxy node returns a response.