Common performance monitoring metrics

Last Updated：2023-08-18 09:26:34 Updated

The following table lists the common performance monitoring metrics on OCP. Note

This section describes monitoring metrics on OCP V2.4.4 as an example. For information about the monitoring metrics on OCP of other versions, see the Monitoring Metrics section in the OCP User Guide of the corresponding version.

Metric group	Metric name	Description	Calculation expression
CPU utilization	cpu_percent	The CPU utilization.	100 * (1 - sum(rate(node_cpu_seconds_total{mode="idle", @LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) / sum(rate(node_cpu_seconds_total{@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS))
IO throughput rate	read	The amount of data read each time.	avg(rate(node_disk_read_bytes_total{@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) / 1048576
IO throughput rate	write	The amount of data written each time.	avg(rate(node_disk_written_bytes_total{@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) / 1048576
I/O time	read	The average read time per second.	1000000 * avg(rate(node_disk_read_time_seconds_total{@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)
I/O time	write	The average write time per second.	1000000 * avg(rate(node_disk_write_time_seconds_total{@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)
IOPS	read	The number of reads per second.	avg(rate(node_disk_reads_completed_total{@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)
IOPS	write	The number of writes per second.	avg(rate(node_disk_writes_completed_total{@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)
Linux system load	load1	The average system load in the last 1 minute.	avg(node_load1{@LABELS}) by (@GBLABELS)
Linux system load	load15	The average system load in the last 15 minutes.	avg(node_load15{@LABELS}) by (@GBLABELS)
Linux system load	load5	The average system load in the last 5 minutes.	avg(node_load5{@LABELS}) by (@GBLABELS)
MEMStore	active	The total size of active memstores.	sum(sysstat_value{metric_group="sysstat",stat_id="130000",@LABELS}) by (@GBLABELS) / 1048576
MEMStore	limit	The upper limit of the total size of all memstores.	sum(sysstat_value{metric_group="sysstat",stat_id="130004",@LABELS}) by (@GBLABELS) / 1048576
MEMStore	total	The total memstore size.	sum(sysstat_value{metric_group="sysstat",stat_id="130001",@LABELS}) by (@GBLABELS) / 1048576
MEMStore	trigger	The threshold that triggers a major compaction.	sum(sysstat_value{metric_group="sysstat",stat_id="130002",@LABELS}) by (@GBLABELS) / 1048576
QPS	all	The number of statements processed per second.	sum(rate(sysstat_value{metric_group="sysstat",stat_id="40002",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) + sum(rate(sysstat_value{metric_group="sysstat",stat_id="40004",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) + sum(rate(sysstat_value{metric_group="sysstat",stat_id="40006",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) + sum(rate(sysstat_value{metric_group="sysstat",stat_id="40008",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) + sum(rate(sysstat_value{metric_group="sysstat",stat_id="40000",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)
QPS	delete	The number of Delete statements processed per second.	sum(rate(sysstat_value{metric_group="sysstat",stat_id="40008",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)
QPS	insert	The number of Insert statements processed per second.	sum(rate(sysstat_value{metric_group="sysstat",stat_id="40002",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)
QPS	replace	The number of Replace statements processed per second.	sum(rate(sysstat_value{metric_group="sysstat",stat_id="40004",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)
QPS	select	The number of Select statements processed per second.	sum(rate(sysstat_value{metric_group="sysstat",stat_id="40000",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)
QPS	update	The number of Update statements processed per second.	sum(rate(sysstat_value{metric_group="sysstat",stat_id="40006",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)
SQL response time	all	The average processing time of each SQL statement on the server.	(sum(rate(sysstat_value{metric_group="sysstat",stat_id="40003",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) + sum(rate(sysstat_value{metric_group="sysstat",stat_id="40005",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) + sum(rate(sysstat_value{metric_group="sysstat",stat_id="40007",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) + sum(rate(sysstat_value{metric_group="sysstat",stat_id="40009",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) + sum(rate(sysstat_value{metric_group="sysstat",stat_id="40001",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)) /(sum(rate(sysstat_value{metric_group="sysstat",stat_id="40002",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) + sum(rate(sysstat_value{metric_group="sysstat",stat_id="40004",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) + sum(rate(sysstat_value{metric_group="sysstat",stat_id="40006",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) + sum(rate(sysstat_value{metric_group="sysstat",stat_id="40008",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) + sum(rate(sysstat_value{metric_group="sysstat",stat_id="40000",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS))
SQL execution plan category	distributed	The number of distributed execution plans processed per second.	sum(rate(sysstat_value{metric_group="sysstat",stat_id="40012",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)
SQL execution plan category	local	The number of local execution plans processed per second.	sum(rate(sysstat_value{metric_group="sysstat",stat_id="40010",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)
SQL execution plan category	remote	The number of remote execution plans processed per second.	sum(rate(sysstat_value{metric_group="sysstat",stat_id="40011",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)
TPS	trans_count	The number of transactions processed per second.	sum(rate(sysstat_value{metric_group="sysstat",stat_id="30005",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)
Event waiting_time	wait_time	The average wait time of a wait event.	sum(rate(time_wait{metric_group="waitevent",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) / sum(rate(total_waits{metric_group="waitevent",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)
Event waiting_number of times	wait_count	The average number of wait events per second.	sum(rate(total_waits{metric_group="waitevent",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)
Transaction response time	trans_time	The average processing time of each transaction on the server.	sum(rate(sysstat_value{metric_group="sysstat",stat_id="30006",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) / sum(rate(sysstat_value{metric_group="sysstat",stat_id="30005",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)
Number of transaction logs	log_count	The number of transaction logs submitted per second.	sum(rate(sysstat_value{metric_group="sysstat",stat_id="30002",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)
Transaction log time-consuming	sync_time	The average time consumed by each synchronization of transaction logs over the network.	sum(rate(sysstat_value{metric_group="sysstat",stat_id="30000",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) / sum(rate(sysstat_value{metric_group="sysstat",stat_id="30001",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)
Transaction log time-consuming	write_disk	The average time required for writing transaction logs to the disk each time.	sum(rate(sysstat_value{metric_group="sysstat",stat_id="80041",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) / sum(rate(sysstat_value{metric_group="sysstat",stat_id="80040",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)
Transaction log volume	log_size	The total size of transaction logs submitted per second.	sum(rate(sysstat_value{metric_group="sysstat",stat_id="80057",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)
Memory	buffers	The size of the kernel buffer cache.	avg(node_memory_Buffers_bytes{@LABELS}) by (@GBLABELS) / 1073741824
Memory	free	The size of the available physical memory.	avg(node_memory_MemFree_bytes{@LABELS}) by (@GBLABELS) / 1073741824
Memory	used	The size of the physical memory used.	(avg(node_memory_MemTotal_bytes{@LABELS}) by (@GBLABELS) - avg(node_memory_MemFree_bytes{@LABELS}) by (@GBLABELS) - avg(node_memory_Cached_bytes{@LABELS}) by (@GBLABELS) - avg(node_memory_Buffers_bytes{@LABELS}) by (@GBLABELS)) / 1073741824
Response time	all	The average processing time of each SQL statement on the server.	(sum(rate(sysstat_value{metric_group="sysstat",stat_id="40003",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) + sum(rate(sysstat_value{metric_group="sysstat",stat_id="40005",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) + sum(rate(sysstat_value{metric_group="sysstat",stat_id="40007",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) + sum(rate(sysstat_value{metric_group="sysstat",stat_id="40009",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) + sum(rate(sysstat_value{metric_group="sysstat",stat_id="40001",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)) /(sum(rate(sysstat_value{metric_group="sysstat",stat_id="40002",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) + sum(rate(sysstat_value{metric_group="sysstat",stat_id="40004",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) + sum(rate(sysstat_value{metric_group="sysstat",stat_id="40006",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) + sum(rate(sysstat_value{metric_group="sysstat",stat_id="40008",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) + sum(rate(sysstat_value{metric_group="sysstat",stat_id="40000",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS))
Response time	delete	The average processing time of each Delete statement on the server.	sum(rate(sysstat_value{metric_group="sysstat",stat_id="40009",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) / sum(rate(sysstat_value{metric_group="sysstat",stat_id="40008",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)
Response time	insert	The average processing time of each Insert statement on the server.	sum(rate(sysstat_value{metric_group="sysstat",stat_id="40003",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) / sum(rate(sysstat_value{metric_group="sysstat",stat_id="40002",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)
Response time	replace	The average processing time of each Replace statement on the server.	sum(rate(sysstat_value{metric_group="sysstat",stat_id="40005",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) / sum(rate(sysstat_value{metric_group="sysstat",stat_id="40004",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)
Response time	select	The average processing time of each Select statement on the server.	sum(rate(sysstat_value{metric_group="sysstat",stat_id="40001",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) / sum(rate(sysstat_value{metric_group="sysstat",stat_id="40000",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)
Response time	update	The average processing time of each Update statement on the server.	sum(rate(sysstat_value{metric_group="sysstat",stat_id="40007",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) / sum(rate(sysstat_value{metric_group="sysstat",stat_id="40006",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)
Capacity_Number of Partitions	partition_count	The number of partitions.	sum(partition_count{metric_group="all_meta_table",@LABELS}) by (@GBLABELS)
Capacity_Number of tables	table_count	The number of tables.	max(table_count{metric_group="all_table",@LABELS}) by (@GBLABELS)
Query response time	all	The average processing time of each SQL statement on the server.	(sum(rate(sysstat_value{metric_group="sysstat",stat_id="40003",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) + sum(rate(sysstat_value{metric_group="sysstat",stat_id="40005",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) + sum(rate(sysstat_value{metric_group="sysstat",stat_id="40007",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) + sum(rate(sysstat_value{metric_group="sysstat",stat_id="40009",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) + sum(rate(sysstat_value{metric_group="sysstat",stat_id="40001",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)) /(sum(rate(sysstat_value{metric_group="sysstat",stat_id="40002",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) + sum(rate(sysstat_value{metric_group="sysstat",stat_id="40004",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) + sum(rate(sysstat_value{metric_group="sysstat",stat_id="40006",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) + sum(rate(sysstat_value{metric_group="sysstat",stat_id="40008",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) + sum(rate(sysstat_value{metric_group="sysstat",stat_id="40000",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS))
Number of active sessions	active_session	The number of active sessions.	sum(active_sessions{metric_group="all_virtual_processlist",@LABELS}) by (@GBLABELS)
Waiting for events	wait_count	The average number of wait events per second.	sum(rate(total_waits{metric_group="waitevent",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)
Waiting for event time	wait_time	The average wait time of a wait event.	sum(rate(time_wait{metric_group="waitevent",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) / sum(rate(total_waits{metric_group="waitevent",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)
Wait for lock time	wait_time	The average wait time of each write lock.	sum(rate(sysstat_value{metric_group="sysstat",stat_id="60023",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) / (sum(rate(sysstat_value{metric_group="sysstat",stat_id="60021",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) + sum(rate(sysstat_value{metric_group="sysstat",stat_id="60022",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS))
Cache hit rate	block_cache	The hit rate of the block cache.	100 * 1 / (1 + sum(rate(sysstat_value{metric_group="sysstat",stat_id="50009",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) / sum(rate(sysstat_value{metric_group="sysstat",stat_id="50008",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS))
Cache hit rate	plan_cache	The hit rate of the plan cache.	100 * sum(rate(hit_count{metric_group="plan_cache_stat",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) / sum(rate(access_count{metric_group="plan_cache_stat",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)
Cache hit rate	row_cache	The hit rate of the row cache.	100 * 1 / (1 + sum(rate(sysstat_value{metric_group="sysstat",stat_id="50001",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) / sum(rate(sysstat_value{metric_group="sysstat",stat_id="50000",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS))
Cache size	block_cache	The size of the block cache.	sum(cache_size{metric_group="all_virtual_kvcache_info",cache_name="user_block_cache",@LABELS}) by (@GBLABELS) / 1048576
Cache size	plan_cache	The size of the plan cache.	sum(mem_used{metric_group="plan_cache_stat",@LABELS}) by (@GBLABELS) / 1048576
Cache size	row_cache	The size of the row cache.	sum(cache_size{metric_group="all_virtual_kvcache_info",cache_name="user_row_cache",@LABELS}) by (@GBLABELS) / 1048576
Network throughput	receive	The amount of data received per second.	avg(rate(node_network_receive_bytes_total{@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) / 1048576
Network throughput	send	The amount of data sent per second.	avg(rate(node_network_transmit_bytes_total{@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) / 1048576
Request waiting queue	queue_count	The average number of SQL statements queued up per second.	sum(rate(sysstat_value{metric_group="sysstat",stat_id="20001",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)
Request waiting queue time-consuming	queue_time	The average waiting time of an SQL statement in a queue.	sum(rate(sysstat_value{metric_group="sysstat",stat_id="20002",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) / sum(rate(sysstat_value{metric_group="sysstat",stat_id="20001",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)
Lock wait	fail	The number of write lock wait failures.	sum(rate(sysstat_value{metric_group="sysstat",stat_id="60022",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)
Lock wait	success	The number of write lock wait successes.	sum(rate(sysstat_value{metric_group="sysstat",stat_id="60021",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS)