PGXC_TOTAL_USER_RESOURCE_INFO
The PGXC_TOTAL_USER_RESOURCE_INFO view displays real-time resource consumption information of users on all instances. This view is supported only by clusters of version 8.2.0 or later.
Name |
Type |
Description |
---|---|---|
nodename |
name |
Instance name, including CNs and DNs. |
username |
name |
Username |
used_memory |
integer |
Used memory (unit: MB) On a DN, it indicates a user's memory usage on the current DN. On a CN, it indicates a user's total memory usage on all DNs. |
total_memory |
integer |
Available memory (unit: MB). 0 indicates that the available memory is not limited and depends on the maximum memory available in the database. On a DN, it indicates the memory available to a user on the current DN. On a CN, it indicates the total memory available to a user on all DNs. |
used_cpu |
double precision |
Number of CPU cores in use. Only the CPU usage of complex jobs in the non-default resource pool is collected, and the value is the CPU usage of the related cgroup. On a DN, it indicates a user's CPU core usage on the current DN. On a CN, it indicates a user's total CPU core usage on all DNs. |
total_cpu |
integer |
Total number of CPU cores of the Cgroups associated with a user. On a DN, it indicates the CPU cores available to a user on the current DN. On a CN, it indicates the total CPU cores available to a user on all DNs. |
used_space |
bigint |
Used permanent table storage space (unit: KB) On a DN, it indicates the size of the permanent table storage space used by a user on the current DN. On a CN, it indicates the total size of the permanent table storage space used by a user on all DNs. |
total_space |
bigint |
Available storage space (unit: KB). -1 indicates that the storage space is not limited. On a DN, it indicates the size of the permanent table storage space available to a user on the current DN. On a CN, it indicates the total size of the permanent table storage space available to a user on all DNs. |
used_temp_space |
bigint |
Used temporary table storage space (unit: KB) On a DN, it indicates the size of the temporary table storage space used by a user on the current DN. On a CN, it indicates the total size of the temporary table storage space used by a user on all DNs. |
total_temp_space |
bigint |
Available temporary table storage space (unit: KB). -1 indicates that the storage space is not limited. On a DN, it indicates the size of the temporary table storage space available to a user on the current DN. On a CN, it indicates the total size of the temporary table storage space available to a user on all DNs. |
used_spill_space |
bigint |
Size of space used for operator spill to disk, in KB. On a DN, it indicates the space used by a user to spill operators to disk on the current DN. On a CN, it indicates the total space used by a user's operators spilled to disk on all DNs. |
total_spill_space |
bigint |
Size of space available for operator spill to disk, in KB. The value -1 indicates that the space is not limited. On a DN, it indicates the space available for a user to spill operators to disk on the current DN. On a CN, it indicates the total space available for a user to spill operators to disk on all DNs. |
read_kbytes |
bigint |
On a CN, it indicates the total number of bytes logically read by a user on all DNs in the last 5 seconds, in KB. On a DN, it indicates the total number of bytes logically read by a user from the instance startup time to the current time, in KB. |
write_kbytes |
bigint |
On a CN, it indicates the total number of bytes logically written by a user on all DNs in the last 5 seconds, in KB. On a DN, it indicates the total number of bytes logically written by a user from the instance startup time to the current time, in KB. |
read_counts |
bigint |
On a CN, it indicates the total number of logical reads performed by a user on all DNs in the last 5 seconds. On a DN, it indicates the total number of logical reads performed by a user from the instance startup time to the current time. |
write_counts |
bigint |
On a CN, it indicates the total number of logical writes performed by a user on all DNs in the last 5 seconds. On a DN, it indicates the total number of logical writes performed by a user from the instance startup time to the current time. |
read_speed |
double precision |
On a CN, it indicates the average logical read rate of a user on a single DN in the last 5 seconds, in KB/s. On a DN, it indicates the average logical read rate of a user on the DN in the last 5 seconds, in KB/s. |
write_speed |
double precision |
On a CN, it indicates the average logical write rate of a user on a single DN in the last 5 seconds, in KB/s. On a DN, it indicates the average logical write rate of a user on the DN in the last 5 seconds, in KB/s. |
send_speed |
double precision |
On a CN, it indicates the sum of the average network sending rates of a user on all DNs in the last 5 seconds, in KB/s. On a DN, it indicates the average network sending rate of a user on the DN in the last 5 seconds, in KB/s. |
recv_speed |
double precision |
On a CN, it indicates the sum of the average network receiving rates of a user on all DNs in the last 5 seconds, in KB/s. On a DN, it indicates the average network receiving rate of a user on the DN in the last 5 seconds, in KB/s. |
Feedback
Was this page helpful?
Provide feedbackThank you very much for your feedback. We will continue working to improve the documentation.See the reply and handling status in My Cloud VOC.
For any further questions, feel free to contact us through the chatbot.
Chatbot