sge_user
Lists all users referenced in the database.
Column | Type | Description |
u_id | Integer | Unique user id |
u_user | text | Name of the user. |
sge_user_values
The sge_user_values table stores the values of user related variables that are
subject to change. These values are currently derived queue values, e.g. hourly averages,
sums etc.
Column | Type | Description |
uv_parent | integer | References q_id in the sge_queue table. |
uv_time_start | timestamp | Start time for the validity of a value. |
uv_time_end | timestamp | End time for the validity of a value. |
uv_variable | text | Variable name, e.g. h_sum_cpu |
uv_value | text | Variable value, e.g. 23.2 |
uv_dvalue | double precision | Variable value as number. |
uv_dconfig | double precision | In case of consumables: Consumable maximum available value (configured value). |
sge_group
Lists all user groups referenced in the database.
Column | Type | Description |
g_id | Integer | Unique group id |
g_group | text | Name of the group. |
Table 15: The sge_group Table
sge_group_values
The sge_group_values table stores the values of group related variables that
are subject to change. These are currently derived values, e.g. hourly averages, sums
etc.
Column | Type | Description |
gv_parent | integer | References q_id in the sge_queue table. |
gv_time_start | timestamp | Start time for the validity of a value. |
gv_time_end | timestamp | End time for the validity of a value. |
gv_variable | text | Variable name, e.g. h_sum_jobs. |
gv_value | text | Variable value, e.g. 53 |
gv_dvalue | double precision | Variable value as number. |
gv_dconfig | double precision | In case of consumables: Consumable maximum available value (configured value). |
sge_daemon_status
Contains information about the running N1GE(EE) daemons.
This information will be delivered by the N1GE(EE) daemons at regular intervals
beginning with version 6.0.
Column | Type | Description |
ds_id | Integer | Unique id for daemon record |
ds_name | text | Name of the daemon, e.g. execd or qmaster. |
ds_hostname | text | Name of the host the daemon in running on. |
ds_pid | integer | The daemons process id. |
ds_curr_time | timestamp | The timestamp of the measurement. |
ds_wallclock | integer | The wallclock time since the startup of the daemon in seconds. |
ds_ru_utime | double | user time used |
ds_ru_stime | double | system time used |
ds_ru_maxrss | integer | maximum resident set size |
ds_ru_idrss | integer | integral resident set size |
ds_ru_minflt | integer | page faults not requiring physical I/O |
ds_ru_majflt | integer | page faults requiring physical I/O |
ds_ru_nswap | integer | swaps |
ds_ru_inblock | integer | block input operations |
ds_ru_oublock | integer | block output operations |
ds_ru_msgsnd | integer | messages sent |
ds_ru_msgrcv | integer | messages received |
ds_ru_nsignals | integer | signals received |
ds_ru_nvcsw | integer | voluntary context switches |
ds_ru_nivcsw | integer | involuntary context switches |
List of Predefined Views
view_accounting
Accounting records for jobs, array tasks, and tightly integrated tasks. Contains
only finished jobs.
Column | Type | Description |
job_number | integer | Job number |
task_number | integer | Array task id |
pe_taskid | text | ID of a tightly integrated parallel task |
name | text | job name (script name or value set with the submit option -N) |
groupname | text | UNIX group name of the primary group the job was executed in. References the
group table. |
username | text | UNIX user account the job was running in. References the user table. |
account | text | Account string set with the submit option -A |
project | text | Project, References the project table |
department | text | Department, References the department table |
submission_time | timestamp | Time of the job submission |
start_time | timestamp | Time when the job was started |
end_time | timestemp | Time when the job finished |
wallclock_time | integer | end_time - start_time |
cpu | double | The CPU time usage in seconds |
io | double | The amount of data transferred in input/outputoperations |
iow | double | The io wait time in seconds |
maxvmem | double | The maximum vmem size in bytes |
wait_time | integer | start_time - submission_time |
turnaround_time | integer | end_time - submission_time |
view_job_times
This is the same as view_accounting, but no tasks of tightly integrated parallel
jobs are listed.
view_jobs_completed
Finished jobs per hour, one record per hour.
Column | Type | Description |
completed | integer | Completed jobs |
time | timestamp | Time when the jobs finished |
view_job_log
Job logging (e.g. Submission, state changes, job finish).
Column | Type | Description |
job_number | integer | Job number |
task_number | integer | Array task id |
pe_taskid | text | ID of a tightly integrated parallel task |
name | text | job name (script name or value set with the submit option -N) |
username | text | UNIX group name of the primary group the job was executed in. References the
group table |
account | text | UNIX user account the job was running in.References the user table |
project | text | Project. References the project table |
department | text | Department. References the department table |
time | timestamp | Time when the job logging entry was generated |
event | text | Event being recorded |
state | text | Job state after the reported event |
initiator | text | User who initiated action for the event |
host | text | Host on with the event action was initiated |
message | text | A message explaining what happened |
|