Updated on 2024-06-03 GMT+08:00

PG_RESOURCE_POOL

PG_RESOURCE_POOL provides information about database resource pools.

Table 1 PG_RESOURCE_POOL columns

Name

Type

Description

oid

oid

Row identifier (hidden attribute, which must be specified).

respool_name

name

Name of the resource pool.

mem_percent

integer

Percentage of the memory configuration

cpu_affinity

bigint

Value of cores bound to the CPU

control_group

name

Name of the Cgroup where the resource pool is located

active_statements

integer

Maximum number of concurrent statements in the resource pool

max_dop

integer

Maximum scanning concurrency during data redistribution. This column is used only for scaling.

memory_limit

name

Maximum memory of the resource pool

parentid

oid

OID of the parent resource pool

io_limits

integer

Upper limit of IOPS. It is counted by ones and by 10 thousands.

io_priority

name

Specifies the I/O priority set for jobs that consume many I/O resources. It takes effect when the I/O usage reaches 90%.

nodegroup

name

Name of the logical database to which the resource pool belongs This column is not supported in the centralized system.

is_foreign

Boolean

Specifies whether the resource pool can be used for users outside the logical database. This column is not supported in the centralized system.

  • true: The resource pool controls the resources of common users who do not belong to the current resource pool.
  • false: The resources of common users who do not belong to the current resource pool are not controlled.

max_worker

integer

Concurrency in a table during data redistribution. This column is used only for scaling.

max_connections

integer

Maximum number of connections that can be used by a resource pool.

max_dynamic_memory

name

Maximum dynamic memory that can be used by a resource pool.

max_shared_memory

name

Maximum shared memory that can be used by a resource pool.

max_concurrency

integer

Maximum number of concurrent requests that can be used by a resource pool.

Note: max_dop and max_worker are used for scaling and are not applicable to the centralized deployment.