HPC infrastructure measures dataset - slurm_job_table table [13/13]

HPC infrastructure measures dataset - slurm_job_table table This record contains the slurm_job_table table of the HPC infrastructure measures dataset. Column descriptions: mod_time: The raw modification time of the record, stored as Unix epoch seconds. timestamp: mod_time converted to a timestamp format. time_submit: The job submission time, stored as Unix epoch seconds. timestamp_submit: time_submit converted to a timestamp format. time_eligible: The time at which the job became eligible to run, stored as Unix epoch seconds. time_start: The job start time, stored as Unix epoch seconds. timestamp_start: time_start converted to a timestamp format. time_end: The job end time, stored as Unix epoch seconds. timestamp_end: time_end converted to a timestamp format. time_suspended: The raw field related to suspension time, stored as Unix epoch seconds. cpus_req: The number of CPUs requested by the job. mem_req: The amount of memory requested by the job. qos_req: The quality of service requested by the job. tres_req: The requested Trackable RESources (TRES) recorded for the job. tres_alloc: The allocated Trackable RESources (TRES) recorded for the job. gres_used: The generic resources used by the job, if tracked. nodes_alloc: The number of nodes allocated to the job. nodelist: The list of nodes allocated to or used by the job. partition: The partition in which the job was submitted or executed. timelimit: The time limit assigned to the job. id_job: The Slurm job identifier. id_assoc: The Slurm association identifier under which the job ran. account: The Slurm account associated with the job (anonymized). priority: The scheduling priority assigned to the job. state: The Slurm state of the job, such as pending, running, completed, failed, or cancelled. state_reason_prev: The previously recorded reason code associated with the job state. flags: Slurm job or scheduling flags describing how the job was handled. restart_cnt: The number of times the job was restarted or requeued. derived_ec: The derived exit code of the job, typically based on the highest exit code returned by its steps. exit_code: The exit code returned by the job. failed_node: The node whose failure caused the job to terminate, when applicable. kill_requid: Internal Slurm field identifying the requester associated with a kill or cancel action, when present. array_task_str: The textual representation of the job-array task specification, when the job belongs to an array. array_max_tasks: The maximum number of array tasks allowed to run concurrently. array_task_pending: Internal job-array information related to pending tasks. id_array_job: The identifier of the parent array job. id_array_task: The identifier of the specific task within the job array. mcs_label: The MCS security label associated with the job. node_inx: Internal Slurm representation of the allocated nodes.