Updated on 2024-09-02 GMT+08:00

GS_RESPOOL_MONITOR

Displays the job running information and resource usage information of all resource pools. This view can be queried only on CNs. This view is supported only by clusters of 8.2.1.100 and later versions.

Table 1 GS_RESPOOL_MONITOR columns

Column

Type

Description

rpname

name

Resource pool name.

nodegroup

name

Name of the logical cluster the resource pool belongs to. The default value is installation.

cn_count

bigint

Number of CNs in the cluster. This parameter is used to determine whether the management and control result of a single CN is proper in a multi-CN environment.

short_acc

boolean

Whether to enable short query acceleration for a resource pool.

session_count

bigint

Number of sessions associated with the resource pool, that is, the number of sessions initiated by users associated with the resource pool, including idle and active sessions.

active_count

bigint

Number of active sessions associated with the resource pool, that is, the number of sessions that are performing queries.

global_wait

bigint

Number of jobs associated with the resource pool that are queued because the number of concurrent jobs on a single CN exceeds the value of max_active_statements.

fast_run

bigint

Number of jobs associated with the resource pool that are running on the fast lane.

fast_wait

bigint

Number of jobs associated with the resource pool that are queued on the fast lane.

fast_limit

bigint

Maximum number of concurrent jobs on the fast lane in a resource pool.

slow_run

bigint

Number of jobs associated with the resource pool that are running on the slow lane.

slow_wait

bigint

Number of jobs associated with the resource pool that are queued on the slow lane.

slow_limit

bigint

Maximum number of concurrent jobs on the slow lane in a resource pool.

used_mem

text

Average memory usage of the resource pool on all DNs. The result has been formatted using pg_size_pretty.

estimate_mem

text

Total estimated memory of jobs running in the resource pool. The result has been formatted using pg_size_pretty.

mem_limit

text

Upper limit of the available memory in the resource pool. The result has been formatted using pg_size_pretty.

query_mem_limit

name

Maximum memory that can be used by a single query in a resource pool. This parameter is used to limit the estimated query memory to prevent abnormal queuing caused by overestimation. The estimated memory is used to limit the actually used query memory. The displayed result has been formatted using pg_size_pretty.

used_cpu

double precision

Average number of CPU cores occupied by a resource pool on all DNs. CPU isolation is performed by node and resource pool. If a single node contains multiple DNs, the number of CPU cores occupied by a resource pool on a single node must be multiplied by the number of DNs.

cpu_limit

double precision

Average upper limit of available CPUs for a resource pool on all nodes. If CPU Time Limit is enabled, the value is the total number of available CPU cores of GaussDB(DWS). If CPU Usage Limit is enabled, the value is the number of available CPU cores of the associated Cgroup.

read_speed

text

Average logical I/O read rate of the resource pool on all DNs. The result has been formatted using pg_size_pretty.

write_speed

text

Average logical I/O write rate of the resource pool on all DNs. The result has been formatted using pg_size_pretty.

send_speed

text

Average network sending rate of the resource pool on all DNs. The result has been formatted using pg_size_pretty.

recv_speed

text

Average receiving rate of the resource pool on all DNs. The result has been formatted using pg_size_pretty.