Common metrics in ARMS Application Monitoring - Application Real-Time Monitoring Service

Use these metrics to build custom Grafana dashboards and PromQL queries for ARMS Application Monitoring.

Important

Applications connected through Managed Service for OpenTelemetry support only business metrics. JVM metrics, system metrics, and other metric categories are not available for these applications.

Metric naming conventions

ARMS metrics follow consistent naming patterns. Understanding these patterns helps you find the right metric and construct queries faster.

Business metrics

Business metrics use the pattern arms_$callType_requests_<suffix>, where $callType is the service access type. Replace $callType with the actual access type when querying.

Example: To query HTTP request counts, use arms_http_requests_count.

For all supported access types, see Service access types and dimensions.

Aggregated business metrics

Aggregated metrics use two naming patterns:

Pattern	Format	Description
Full-dimension	`<prefix>_raw`	Includes all observation dimensions
Reduced-dimension	`<prefix>_ign_x_y`	Excludes dimensions `x` and `y` from the metric

Reduced-dimension metrics improve query performance by pre-excluding unused dimensions. Choose the variant with only the dimensions your query requires.

How ARMS Gauge metrics differ from open-source frameworks

Important

Except for quantile and Counter-type metrics, ARMS metrics are Gauge type. Each data point represents the cumulative total within the 15-second collection interval, not an instantaneous value. This affects how you write PromQL expressions.

Scenario	ARMS PromQL	Open-source equivalent
Average QPS over 1 minute	`sum_over_time(arms_$callType_requests_count[1m]) / 60`	`rate(http_server_requests_count[1m])`
Error rate over 5 minutes	`sum_over_time(arms_http_requests_error_count[5m]) / sum_over_time(arms_http_requests_count[5m])`	`rate(http_server_requests_error_count[5m]) / rate(http_server_requests_count[5m])`
P99 latency	`arms_http_requests_latency_seconds{quantile="0.99"}`	`histogram_quantile(0.99, rate(http_server_requests_seconds_bucket[5m]))`

Metric storage locations

Metrics are stored in one of two metric libraries. Specify the correct metricstore when querying.

Storage location	Metricstore name	Metric categories
Core metric library	`metricstore-apm-metrics`	JVM, system, thread pool/connection pool, Go runtime, and the following aggregated business metrics: `arms_app_requests_{metric}_ign_destid_endpoint_parent_ppid_prpc_rpc`, `arms_app_requests_{metric}_ign_destid_endpoint_rpc_prpc`, `arms_app_requests_{metric}_ign_destid_endpoint_parent_ppid_prpc`
Detailed metric library	`metricstore-apm-metrics-detail`	Business metrics, scheduled task metrics, and all other aggregated business metrics

Business metrics

Business metrics track request count, errors, latency, and slow requests per service access type.

Common dimensions

Dimension	Key
Service name	`service`
Service PID	`pid`
Machine IP	`serverIp`
Interface	`rpc`

Metric list

All metrics use a 15-second collection interval.

Metric name	Metric	Type	Unit	Dimensions
Request count	`arms_$callType_requests_count`	Gauge	--	Varies by access type. See Service access types and dimensions.
Error request count	`arms_$callType_requests_error_count`	Gauge	--	Same as above
Request duration	`arms_$callType_requests_seconds`	Gauge	Seconds	Same as above
Slow request count	`arms_$callType_requests_slow_count`	Gauge	--	Same as above
Request duration quantiles	`arms_$callType_requests_latency_seconds`	Summary	Seconds	Available only for HTTP access type with quantile statistics enabled. Quantile values: 0.5, 0.75, 0.90, 0.99. To enable quantile statistics, see Advanced Settings.

Aggregated business metrics

Per-access-type business metrics create long PromQL expressions when an application uses multiple access types (such as HTTP and Dubbo). They also include all observation dimensions, which can degrade query performance when you need only a subset.

Aggregated business metrics solve both problems: they combine access types into unified metrics and offer reduced-dimension variants for faster queries.

Unless otherwise specified, all aggregated business metrics are Gauge type with a 15-second collection interval.

Common dimensions

Every aggregated metric includes these dimensions:

Dimension	Key	Description
Application PID	`pid`	Process identifier
Application name	`service`	Service name
Instance IP	`serverIp`	Host IP address
Metric source	`source`	`apm` = ARMS agent, `xtrace` = Managed Service for OpenTelemetry, `ebpf` = eBPF agent

Category	Tracks	Scope
General	Request count, error count, slow request count, request duration	All access types
Database	Request count, error count, slow request count, request duration	Database access types only
SQL	Same as Database, plus SQL statement dimension	Database access types only
Exception	Request count, request duration	All access types, only when exceptions occur
Status code	Request count by HTTP status code	HTTP access type only
Quantile	Request duration quantiles	All access types (agent V4.x+)

General metrics

Additional dimensions for _raw metrics:

Dimension	Key	Description
Access type	`callType`	HTTP, Dubbo, database, etc.
Interface name	`rpc`	Endpoint or method name
Upstream interface	`prpc`	Caller interface
Upstream application PID	`ppid`	Caller process ID
Upstream application name	`parent`	Caller service name
Peer address	`endpoint`	For `http_client`: peer interface. For other types: remote address.
Target identifier	`destId`	For database types: database name. For other types: remote address.

Request count

Metric	Excluded dimensions
`arms_app_requests_count_raw`	None (all dimensions included)
`arms_app_requests_count_ign_destid_endpoint_parent_ppid_prpc_rpc`	destId, endpoint, parent, ppid, prpc, rpc
`arms_app_requests_count_ign_destid_endpoint_rpc_prpc`	destId, endpoint, rpc, prpc
`arms_app_requests_count_ign_destid_endpoint_parent_ppid_prpc`	destId, endpoint, parent, ppid, prpc
`arms_app_requests_count_ign_parent_ppid_prpc_rpc`	parent, ppid, prpc, rpc

Error request count

Metric	Excluded dimensions
`arms_app_requests_error_count_raw`	None (all dimensions included)
`arms_app_requests_error_count_ign_destid_endpoint_rpc_prpc`	destId, endpoint, rpc, prpc
`arms_app_requests_error_count_ign_destid_endpoint_parent_ppid_prpc`	destId, endpoint, parent, ppid, prpc
`arms_app_requests_error_count_ign_parent_ppid_prpc_rpc`	parent, ppid, prpc, rpc

Slow request count

Metric	Excluded dimensions
`arms_app_requests_slow_count_raw`	None (all dimensions included)
`arms_app_requests_slow_count_ign_destid_endpoint_parent_ppid_prpc_rpc`	destId, endpoint, parent, ppid, prpc, rpc
`arms_app_requests_slow_count_ign_destid_endpoint_rpc_prpc`	destId, endpoint, rpc, prpc
`arms_app_requests_slow_count_ign_destid_endpoint_parent_ppid_prpc`	destId, endpoint, parent, ppid, prpc
`arms_app_requests_slow_count_ign_parent_ppid_prpc_rpc`	parent, ppid, prpc, rpc

Request duration

Metric	Unit	Excluded dimensions
`arms_app_requests_seconds_raw`	Seconds	None (all dimensions included)
`arms_app_requests_seconds_ign_destid_endpoint_parent_ppid_prpc_rpc`	Seconds	destId, endpoint, parent, ppid, prpc, rpc
`arms_app_requests_seconds_ign_destid_endpoint_rpc_prpc`	Seconds	destId, endpoint, rpc, prpc
`arms_app_requests_seconds_ign_destid_endpoint_parent_ppid_prpc`	Seconds	destId, endpoint, parent, ppid, prpc
`arms_app_requests_seconds_ign_parent_ppid_prpc_rpc`	Seconds	parent, ppid, prpc, rpc

Database metrics

Additional dimensions:

Dimension	Key	Description
Access type	`callType`	Database type (MySQL, PostgreSQL, etc.)
Interface name	`rpc`	Query method
Database instance address	`endpoint`	Connection endpoint
Database name	`destId`	Target database

Metric name	Full-dimension metric	Reduced-dimension metric (excludes rpc)
Request count	`arms_db_requests_count_raw`	`arms_db_requests_count_ign_rpc`
Error request count	`arms_db_requests_error_count_raw`	`arms_db_requests_error_count_ign_rpc`
Slow request count	`arms_db_requests_slow_count_raw`	`arms_db_requests_slow_count_ign_rpc`
Request duration	`arms_db_requests_seconds_raw`	`arms_db_requests_seconds_ign_rpc`

SQL metrics

SQL metrics extend database metrics by adding the SQL statement dimension.

Additional dimensions: Same as Database metrics, plus:

Dimension	Key	Description
SQL statement ID	`sqlId`	Identifier for the SQL statement

Metric name	Full-dimension metric	Reduced-dimension metric (excludes rpc)
Request count	`arms_sql_requests_count_raw`	`arms_sql_requests_count_ign_rpc`
Error request count	`arms_sql_requests_error_count_raw`	`arms_sql_requests_error_count_ign_rpc`
Slow request count	`arms_sql_requests_slow_count_raw`	`arms_sql_requests_slow_count_ign_rpc`
Request duration	`arms_sql_requests_seconds_raw`	`arms_sql_requests_seconds_ign_rpc`

Exception metrics

Additional dimensions:

Dimension	Key	Description
Access type	`callType`	Service access type
Interface name	`rpc`	Endpoint or method name
Peer address	`endpoint`	For `http_client`: peer interface. For other types: remote address.
Target identifier	`destId`	For database types: database name. For other types: remote address.
Exception encoding method	`excepInfo`	Encoding format
Exception encoding ID	`excepType`	Unique exception identifier
Exception name	`excepName`	Exception class name

Metric name	Metric	Excluded dimensions
Exception request count	`arms_exception_requests_count_raw`	None
	`arms_exception_requests_count_ign_destid_endpoint_rpc`	destId, endpoint, rpc
	`arms_exception_requests_count_ign_destid_endpoint`	destId, endpoint
Exception request duration	`arms_exception_requests_seconds_raw`	None
	`arms_exception_requests_seconds_ign_destid_endpoint_rpc`	destId, endpoint, rpc
	`arms_exception_requests_seconds_ign_destid_endpoint`	destId, endpoint

Status code metrics

Additional dimensions:

Dimension	Key	Description
Interface name	`rpc`	HTTP endpoint
Status code	`status`	`200`, `2xx` (201-299), `3xx` (300-399), `4xx` (400-499), `5xx` (500+)

Metric name	Metric	Excluded dimensions
Request count by status code	`arms_requests_by_status_count_raw`	None
	`arms_requests_by_status_count_ign_rpc`	rpc

Quantile metrics

Quantile metrics require agent V4.x or later.

Additional dimensions:

Dimension	Key	Description
Access type	`callType`	Service access type
Interface name	`rpc`	Endpoint or method name
Quantile	`quantile`	`0.5` (p50), `0.75` (p75), `0.90` (p90), `0.99` (p99)
Peer address	`endpoint`	For `http_client`: peer interface. For other types: remote address.
Target identifier	`destId`	For database types: database name. For other types: remote address.
Exception encoding method	`excepInfo`	Encoding format
Exception encoding ID	`excepType`	Unique exception identifier
Exception name	`excepName`	Exception class name
Status code	`status`	`200`, `2xx`, `3xx`, `4xx`, `5xx`

Metric	Type
`arms_uni_requests_latency_seconds`	Summary

Choose the right metric

Use the following examples to select the best metric variant for your query.

Example 1: Count requests across all interfaces

The General category provides request count metrics.
Your query needs the interface dimension (rpc) but not upstream, peer, or database dimensions.
Pick the variant that retains rpc while excluding the most unnecessary dimensions.

Best choice: arms_app_requests_count_ign_destid_endpoint_parent_ppid_prpc

This metric keeps rpc (interface) and callType while excluding destId, endpoint, parent, ppid, and prpc, giving you the right data with optimal query performance.

Example 2: Calculate error rate by service

You need error count and total request count at the service level -- no per-interface breakdown.
Use the most reduced-dimension variants for both metrics.

sum_over_time(arms_app_requests_error_count_ign_destid_endpoint_parent_ppid_prpc_rpc[5m])
/
sum_over_time(arms_app_requests_count_ign_destid_endpoint_parent_ppid_prpc_rpc[5m])

Example 3: Monitor slow database queries

The Database category tracks slow requests.
You need the database name (destId) and endpoint, but not the query method (rpc).

Best choice: arms_db_requests_slow_count_ign_rpc

JVM metrics

All JVM metrics use a 15-second collection interval.

Common dimensions

Dimension	Key
Service name	`service`
Service PID	`pid`
Machine IP	`serverIp`

Garbage collection

Metric name	Metric	Type	Unit
Total GC count	`arms_jvm_gc_total`	Counter	--
Total GC duration	`arms_jvm_gc_seconds_total`	Counter	Seconds
GC count per interval	`arms_jvm_gc_delta`	Gauge	--
GC duration per interval	`arms_jvm_gc_seconds_delta`	Gauge	Seconds

Dimensions:

Generational GC area: Young (young generation), Old (old generation)
Cause (agent V4.4.0+): The GC trigger reason. Common causes include:

Cause	Description
`Allocation Failure`	JVM could not allocate memory for a new object, indicating memory pressure
`System.gc()`	Triggered by an explicit `System.gc()` call in application code
`Heap Dump Initiated GC`	Triggered during a heap dump operation
`Metadata GC Threshold`	Triggered when metaspace usage reaches the threshold
`G1 Evacuation Pause`	G1 collector evacuation pause
`G1 Humongous Allocation`	G1 collector humongous object allocation
`No GC`	JVM triggered GC based on internal heuristics (such as CMS heap usage prediction) rather than a specific condition

All GC causes

System.gc()
FullGCAlot / ScavengeAlot
Allocation Profiler
JvmtiEnv ForceGarbageCollection
Heap Inspection Initiated GC / Heap Dump Initiated GC
WhiteBox Initiated Young GC / WhiteBox Initiated Full GC / Run to Breakpoint
No GC
Allocation Failure
CodeCache GC Threshold / Aggressive
Metadata GC Threshold / Clear Soft References
G1 Evacuation Pause / G1 Compaction Pause / G1 Humongous Allocation / G1 Periodic Collection
Diagnostic Command
Shenandoah GC causes
ZGC causes (Timer, Warmup, Allocation Rate, Proactive, etc.)
ILLEGAL VALUE

Threads

Metric name	Metric	Type	Unit
JVM thread count	`arms_jvm_threads_count`	Gauge	--

Dimension -- State:

Blocked, Live, Daemon, New, Dead-lock, Runnable, Terminated, Timed-wait, Wait

Memory

Metric name	Metric	Type	Unit
Initial memory size	`arms_jvm_mem_init_bytes`	Gauge	Bytes
Maximum memory size	`arms_jvm_mem_max_bytes`	Gauge	Bytes
Used memory size	`arms_jvm_mem_used_bytes`	Gauge	Bytes
Committed memory size	`arms_jvm_mem_committed_bytes`	Gauge	Bytes
Memory usage ratio	`arms_jvm_mem_usage_ratio`	Gauge	Ratio (0-1)

Dimensions:

Area: Heap, Nonheap, Total
ID (region breakdown): Eden, Old, Survivor, Metaspace, Code cache, Compressed class space, Total

Classes

Metric name	Metric	Type	Unit
Loaded classes	`arms_class_load_loaded`	Counter	--
Unloaded classes	`arms_class_load_un_loaded`	Counter	--

Buffer pools

Metric name	Metric	Type	Unit
Buffer pool total size	`arms_jvm_buffer_pool_total_bytes`	Gauge	Bytes
Buffer pool used size	`arms_jvm_buffer_pool_used_bytes`	Gauge	Bytes
Buffer pool count	`arms_jvm_buffer_pool_count`	Gauge	--

Dimension -- ID: Direct, Mapped

File descriptors

Metric name	Metric	Type	Unit
Open file descriptors	`arms_file_desc_open_count`	Gauge	--
Open file descriptor ratio	`arms_file_desc_open_ratio`	Gauge	Ratio (0-1)

System metrics

All system metrics are Gauge type with a 15-second collection interval.

Common dimensions

Dimension	Key
Service name	`service`
Service PID	`pid`
Machine IP	`serverIp`

CPU

Metric name	Metric	Unit
CPU idle	`arms_system_cpu_idle`	Percentage
CPU I/O wait	`arms_system_cpu_io_wait`	Percentage
CPU system	`arms_system_cpu_system`	Percentage
CPU user	`arms_system_cpu_user`	Percentage
System load (1 min)	`arms_system_load`	--

Disk

Metric name	Metric	Unit
Free disk space	`arms_system_disk_free_bytes`	Bytes
Total disk space	`arms_system_disk_total_bytes`	Bytes
Disk usage ratio	`arms_system_disk_used_ratio`	Ratio (0-1)

Memory

Metric name	Metric	Unit
Buffer size	`arms_system_mem_buffers_bytes`	Bytes
Cache size	`arms_system_mem_cached_bytes`	Bytes
Free memory	`arms_system_mem_free_bytes`	Bytes
Free swap	`arms_system_mem_swap_free_bytes`	Bytes
Total swap	`arms_system_mem_swap_total_bytes`	Bytes
Total memory	`arms_system_mem_total_bytes`	Bytes
Used memory	`arms_system_mem_used_bytes`	Bytes

Network

Metric name	Metric	Unit
Inbound traffic	`arms_system_net_in_bytes`	Bytes
Outbound traffic	`arms_system_net_out_bytes`	Bytes
Inbound errors	`arms_system_net_in_err`	--
Outbound errors	`arms_system_net_out_err`	--

Thread pool and connection pool metrics

Metric names and dimensions differ between agent versions. Check your agent version and refer to the matching section.

Agent V4.1.x and later

All metrics use a 15-second collection interval.

Thread pool metrics

Metric name	Metric	Type
Core thread count	`arms_thread_pool_core_pool_size`	Gauge
Maximum thread count	`arms_thread_pool_max_pool_size`	Gauge
Active thread count	`arms_thread_pool_active_thread_count`	Gauge
Current thread count	`arms_thread_pool_current_thread_count`	Gauge
Peak thread count	`arms_thread_pool_max_thread_count`	Gauge
Scheduled task count	`arms_thread_pool_scheduled_task_count`	Counter
Completed task count	`arms_thread_pool_completed_task_count`	Counter
Rejected task count	`arms_thread_pool_rejected_task_count`	Counter
Task queue size	`arms_thread_pool_queue_size`	Gauge

Dimensions:

Dimension	Key	Example
Thread name pattern	`thread_name_pattern`	`http-nio-8080-*`
Thread pool purpose	`thread_pool_usage`	Tomcat, Dubbo, Undertow

Connection pool metrics

Metric name	Metric	Type
Connections	`arms_connection_pool_connection_count`	Gauge
Minimum idle connections	`arms_connection_pool_connection_min_idle_count`	Gauge
Maximum idle connections	`arms_connection_pool_connection_max_idle_count`	Gauge
Maximum connections	`arms_connection_pool_connection_max_count`	Gauge
Pending connection requests	`arms_connection_pool_pending_request_count`	Counter

Dimensions:

Dimension	Key	Description
Connection state	`state`	`used` or `idle` (applies to `arms_connection_pool_connection_count` only)
Connection pool type	`pool_type`	Druid, c3p0, etc.
Database connection string	`url`	JDBC connection URL

Agent versions earlier than V4.1.x

If you use agent V4.1.x or later, refer to the previous section instead.

All metrics are Gauge type with a 15-second collection interval.

Common dimensions: service, pid, serverIp, name (thread pool name), type (thread pool type)

Metric name	Metric
Core pool size	`arms_threadpool_core_size`
Maximum pool size	`arms_threadpool_max_size`
Active thread count	`arms_threadpool_active_size`
Queue size	`arms_threadpool_queue_size`
Current pool size	`arms_threadpool_current_size`
Task count by state	`arms_threadpool_task_total`

Dimension for arms_threadpool_task_total -- Status: Scheduled, Completed, Rejected

Scheduled task metrics

Available only for scheduled tasks.

All metrics are Gauge type with a 15-second collection interval.

Common dimensions

Dimension	Key
Service name	`service`
Service PID	`pid`
Machine IP	`serverIp`
Task ID	`rpc`

Metric list

Metric name	Metric	Unit
Scheduling delay	`arms_$callType_delay_milliseconds`	Milliseconds

Replace $callType with the scheduled task type: xxl_job, spring_scheduled, quartz, elasticjob, jdk_timer, or schedulerx.

Go runtime metrics

All Go runtime metrics use a 15-second collection interval.

Common dimensions

Dimension	Key
Service name	`service`
Service PID	`pid`
Machine IP	`serverIp`

Metric list

Metric name	Metric	Type
Application uptime	`arms_golang_runtime_uptime`	Int64Counter
Goroutine count	`arms_golang_process_runtime_go_goroutines`	Gauge
Heap allocated memory	`arms_golang_process_runtime_go_mem_heap_alloc`	Gauge
Idle heap memory	`arms_golang_process_runtime_go_mem_heap_idle`	Gauge
In-use heap memory	`arms_golang_process_runtime_go_mem_heap_inuse`	Gauge
Heap object count	`arms_golang_process_runtime_go_mem_heap_objects`	Gauge
Released heap memory	`arms_golang_process_runtime_go_mem_heap_released`	Gauge
System-requested virtual memory	`arms_golang_process_runtime_go_mem_heap_sys`	Gauge
Live object count	`arms_golang_process_runtime_go_mem_live_objects`	Gauge
GC cycle count	`arms_golang_process_runtime_go_gc_count`	Gauge
Cumulative STW GC pause time	`arms_golang_process_runtime_go_gc_pause_total_ns`	Int64Counter
GC pause time distribution	`arms_golang_process_runtime_go_gc_pause_ns`	Int64Histogram

Service access types and dimensions

When querying business metrics (arms_$callType_requests_*), replace $callType with one of the access types listed below. Each category has its own set of available dimensions.

Client-side types

Access types: http_client, dubbo_client, hsf_client, dsf_client, notify_client, grpc_client, thrift_client, sofa_client, mq_client, kafka_client

Dimension	Key	Description
Upstream service name	`parent`	Caller service
Upstream service PID	`ppid`	Caller process ID
Peer extended info	`destId`	Additional request peer metadata
Peer address	`endpoint`	Request peer address
Exception ID	`excepType`	Exception identifier
Exception encoding rule	`excepInfo`	Exception ID encoding format
Exception name	`excepName`	Exception class name
Stack trace ID	`stackTraceId`	Trace identifier for stack analysis

Database types

Access types: mysql, oracle, mariadb, postgresql, ppas, sqlserver, mongodb, dmdb

Dimension	Key	Description
Upstream service name	`parent`	Caller service
Upstream service PID	`ppid`	Caller process ID
Database name	`destId`	Target database
Database address	`endpoint`	Connection endpoint
Exception ID	`excepType`	Exception identifier
Exception encoding rule	`excepInfo`	Exception ID encoding format
Exception name	`excepName`	Exception class name
Stack trace ID	`stackTraceId`	Trace identifier for stack analysis
SQL statement ID	`sqlId`	SQL identifier

Server-side types

Access types: http, dubbo, hsf, dsf, user_method, mq, kafka, grpc, thrift, sofa

Dimension	Key	Description
Upstream interface	`prpc`	Caller interface
Upstream service name	`parent`	Caller service
Upstream service PID	`ppid`	Caller process ID
Service address	`endpoint`	Address of the service
Exception ID	`excepType`	Exception identifier
Exception encoding rule	`excepInfo`	Exception ID encoding format
Exception name	`excepName`	Exception class name
Stack trace ID	`stackTraceId`	Trace identifier for stack analysis

Scheduled task types

Access types: xxl_job, spring_scheduled, quartz, elasticjob, jdk_timer, schedulerx

Dimension	Key	Description
Upstream interface	`prpc`	Caller interface
Upstream service name	`parent`	Caller service
Upstream service PID	`ppid`	Caller process ID
Exception ID	`excepType`	Exception identifier
Exception encoding rule	`excepInfo`	Exception ID encoding format
Exception name	`excepName`	Exception class name
Stack trace ID	`stackTraceId`	Trace identifier for stack analysis

Metric naming conventions

Business metrics

Aggregated business metrics

How ARMS Gauge metrics differ from open-source frameworks

Metric storage locations

Business metrics

Common dimensions

Metric list

Aggregated business metrics

Common dimensions

Categories

General metrics

Request count

Error request count

Slow request count

Request duration

Database metrics

SQL metrics

Exception metrics

Status code metrics

Quantile metrics

Choose the right metric

JVM metrics

Common dimensions

Garbage collection

Threads

Memory

Classes

Buffer pools

File descriptors

System metrics

Common dimensions

CPU

Disk

Memory

Network

Thread pool and connection pool metrics

Agent V4.1.x and later

Thread pool metrics

Connection pool metrics

Agent versions earlier than V4.1.x

Scheduled task metrics

Common dimensions

Metric list

Go runtime metrics

Common dimensions

Metric list

Service access types and dimensions

Client-side types

Database types

Server-side types

Scheduled task types

References