The WI_JOBGROUP table stores the jobGroup field and the user_group_code assigned to the job.
For a new event, the ETL splits the starting index and ending index and inserts a new record during each iteration between the two.
For a modify event, the ETL updates the existing record with the new jobgroup and usergroup values.
The following table describes each column and its source:
Column name |
Source and description |
Key |
CLUSTER_CODE |
This is the cluster code. The source is the ClusterName field in the data file. The cluster code is obtained by looking up the ClusterName in the wi_clustercode table and generating a new code if it doesn't exist. |
Primary key |
QUEUE_TIME |
This is the queue time, in GMT. The source is the LSBEvent field in the data file. The queue time is the SubmitTime subfield, which is split from the LSBEvent field. |
Primary key |
JOB_ID |
This is the job ID. The source is the LSBEvent field in the data file. The job ID is the Job_ID subfield, which is split from the LSBEvent field. |
Primary key |
JOB_ARRAY_INDEX |
This is the job array index. The source is the LSBEvent field in the data file. The job array index consists of the start_index and end_index subfields, which are split from the LSBEvent field. This comes from the "LSBEvent" field in the data file. We split out the start_index and end_index and we assign the index in between and insert a record for each index. |
Primary key |
JOB_GROUP |
This field contains the jobgroup usage options. The source is the LSBEvent field in the data file. If the job has no jobgroup usage options of using the jobgroup, this field is "UNKNOWN". |
|
USER_GROUP_CODE |
This is the user group code. The source is the LSBEvent field in the data file. The User_Group subfield is split out from the LSBEvent field. The user group code is obtained by looking up the User_Group in the wi_usergroupcode table and generating a new code if it doesn't exist. |
|