Win2008 the management job queue for HPC cluster job management

Source: Internet
Author: User
Tags commit error handling

In job management, you can monitor and manage jobs that are submitted to a cluster. In the job list, each row is a job, and each column displays the job properties, job status, and indicator values. The job list provides a starting point for drilling down to job details and performing operations on one or more jobs.

The HPC Cluster Administrator provides several charts and reports to track job statistics for the cluster.

Configure scheduling policies

The scheduling policy determines the order in which jobs are run in the queue. When you configure the HPC Job Scheduler service, you can set scheduling policies such as preemptive and backfill, error handling, and job history options. By creating a custom job submission filter and job activation filter, you can enforce site-specific job submission policies and job activation policies.

View Jobs and tasks

The job list displays information about jobs in the cluster. You can filter and sort the job list, and you can select the job properties and metric values to display in the list. When you click a job in the job list, detailed information about the job appears in the details pane. In the details pane, you can view the tasks, task status, job details, and activity logs for jobs in the job. You can also view job and task results.

1. Configure the HPC Job Scheduler service

The HPC Job Scheduler service can queue jobs and tasks, assign resources, assign tasks to compute nodes, and monitor the status of jobs, tasks, and nodes.

The scheduling policy determines the order in which jobs are run in the queue. When you configure the HPC Job Scheduler service, you can set scheduling policies such as preemptive and backfill, error handling, and job history options. You can also specify custom job activation and submit filter programs.

Configuring the HPC Job Scheduler service

On the Options menu, click Job Scheduler Configuration.

Use the Job Scheduler Configuration dialog box to set schedule options.

When you are done, click Apply to save your changes, or click OK to save and exit the dialog box.

2. Learn about activating and submitting filters

By creating a custom filter, you can enforce site-specific job submission policies and job activation policies. They are called job commit filters and job activation filters.

Job Submission Filter

A job submission filter is a program (. exe or. cmd file) that can be accessed from the head node of the cluster that is running the HPC Job Scheduler service each time the job is submitted. Job submission filters can reject, accept, or change jobs that users submit to the cluster. The job submission filter analyzes the job description file to see if there are prohibited options or if the required options are not included. For example, you might want to disable the default Run-time value of Infinite. This can be done through a commit filter that can check the job description file for "Infinite" and, if so, change it to a finite value.

Attention

The commit filter allows you to change only the job property values. You cannot use the Submit filter to change task property options.

In another example, you might want to require a new job option created by editing the job schema. They are called extended options, and the scheduler does not check them. Therefore, if you want to enforce the requirement and prevent the job from entering the queue if the option is not found, you must use a submit filter.

Job Activation Filter

A Job activation filter is a program (. exe or. cmd file) that can be accessed from the head node of the cluster that is running the HPC Job Scheduler service whenever the job is ready to start and can delay or allow the job to start. The job activation filter checks whether there are factors in the queued job that could cause the job to fail if the activation job occurs, such as when the license is unavailable or exceeds the time used by the submitting user. If the filter detects this condition, it prevents the job from being activated.

Contact Us

The content source of this page is from Internet, which doesn't represent Alibaba Cloud's opinion; products and services mentioned on that page don't have any relationship with Alibaba Cloud. If the content of the page makes you feel confusing, please write us an email, we will handle the problem within 5 days after receiving your email.

If you find any instances of plagiarism from the community, please send an email to: info-contact@alibabacloud.com and provide relevant evidence. A staff member will contact you within 5 working days.

A Free Trial That Lets You Build Big!

Start building with 50+ products and up to 12 months usage for Elastic Compute Service

  • Sales Support

    1 on 1 presale consultation

  • After-Sales Support

    24/7 Technical Support 6 Free Tickets per Quarter Faster Response

  • Alibaba Cloud offers highly flexible support services tailored to meet your exact needs.