doc-exports/docs/dataartsstudio/umn/dataartsstudio_01_0506.html

Overview

Choose Monitoring > Overview. On the Overview page, you can view the statistics of job instances in charts. Currently, you can view seven types of statistics:

• Status
  • You can filter out your or all job instances by selecting My Instances or All, and filter out job instances of the current day by selecting Today.
  • You can filter out your or all job instances by selecting My Instances or All, and filter out job instances of the current day by selecting Yesterday.
  • You can filter out your or all job instances by selecting My Instances or All, and filter out job instances of the current day by selecting Two days ago.
  • You can filter out your or all job instances by selecting My Instances or All, and filter out job instances of the current day by selecting Last 7 days.
  • You can click a status to go to the Monitor Instance page and view details about all jobs in the status.
    

    • The statistics include the monitoring data of the instances of real-time jobs. When you click a status, you will not be redirected to the Monitor Instance page of real-time jobs. Instead, you can only view the details of the instances of batch jobs.
    • By default, the system displays all job instances of the current day.
    • You can view the total number of instances that meet the filter criteria, the total number of instances that were successfully executed, and the percentage of such instances.
      Figure 1 Status
      
• Completed Tasks
  

  Successfully executed instances of the current day are collected once an hour. A task is a job node.

  • You can specify a date and view the number of all job nodes successfully executed on the previous day, on the selected day, and over the last seven days on average.
  • You can specify a date and view the number of different types of job nodes successfully executed on the previous day, on the selected day, and over the last seven days on average.
• Tasks
  

  Number of tasks (operators in jobs) started in five minutes. Data of 30 days is available.

  • You can filter the operators started on each day within 30 days.
  • You can view the curve of the number of all operators that have been started.
  • You can view the curve of the number of different types of operators that have been started.
• Running DLI Jobs/Queue CU Usage

  You can filter the number of running DLI jobs and the CU usage of a specified queue.

  

  • You can view data of the last seven days by default, and view data of one month at most.
  • You can only view data of non-default queues. You can click the name of a queue to pin the queue to top.
• Number of Jobs and Number of Tasks Scheduled Daily
  

  This area displays the trend of the total number of jobs in a long period and the number of tasks scheduled each day. A task indicates an operator in a job.

  Number of jobs: total number of batch processing jobs and real-time jobs

  Number of tasks scheduled daily: number of tasks scheduled everyday, including both real-time and offline tasks. The number is calculated based on the nodes that are successfully scheduled.

  • By default, the system displays the number of jobs and the number of tasks scheduled each day in one month. You can filter data by time range.
• Task Type Distribution

  You can view the number of job nodes of different types.

  

  A task indicates an operator in a job.

  The system collects statistics on the number of nodes in all submitted jobs, including real-time jobs and batch processing jobs.

• Top 100 in Instance Running Time
  • You can filter out the top 100 instances of yours or all users with the longest running duration by time and owner.
  • You can click a job name to go to the Monitor Instance page and view the job running details.
  • By default, the system displays top 100 job instances in one month.
• Top 100 in Instance Failed to Run
  • You can filter out the top 100 instances of yours or all users with the most failures by time and owner.
  • You can click a job name to go to the Monitor Instance page, view the logs of the failed job instances, and analyze the causes.
  • By default, the system displays top 100 job instances in one month.
• End of scheduling in the next week

  You can view the jobs that are expected to be completed in the following week, including their names, owners, and end time.

  

  • Jobs that are expected to be completed in two days or less are displayed in red.
  • Jobs that are expected to be completed in three to five days are displayed in orange.
  • Jobs that are expected to be completed in six to seven days are displayed in black.

Parent topic: O&M and Scheduling

<script language="JavaScript"> </script>