You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@airavata.apache.org by "ASF subversion and git services (Jira)" <ji...@apache.org> on 2022/09/08 21:39:00 UTC

[jira] [Commented] (AIRAVATA-3649) Automatically calculate queue settings with custom code

    [ https://issues.apache.org/jira/browse/AIRAVATA-3649?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17602020#comment-17602020 ] 

ASF subversion and git services commented on AIRAVATA-3649:
-----------------------------------------------------------

Commit d273a78699c0b7a0f8687eadcdfc3cb7f998cdec in airavata-django-portal's branch refs/heads/develop from Marcus Christie
[ https://gitbox.apache.org/repos/asf?p=airavata-django-portal.git;h=d273a786 ]

AIRAVATA-3649 Rerun validation in QueueSettingsEditor whenever values change


> Automatically calculate queue settings with custom code
> -------------------------------------------------------
>
>                 Key: AIRAVATA-3649
>                 URL: https://issues.apache.org/jira/browse/AIRAVATA-3649
>             Project: Airavata
>          Issue Type: New Feature
>          Components: Django Portal
>            Reporter: Marcus Christie
>            Assignee: Marcus Christie
>            Priority: Major
>
> For some applications the ideal queue settings for an experiment can be straightforwardly calculated from the input values for the experiment. An administrator may want to define some custom code to calculate the ideal queue settings and then register it for use with one or more applications. This will make streamline the use of the gateway for users and also reduce errors in over- or under-allocating resources.
> h3. Design
> - custom code to be implemented as a module function that takes experiment_model and returns a dictionary of possibly different queue setting values
> - decorator {{@queue_settings_calculator}}
> - decorator attributes: id (allows renaming function), name (display to user)
> - decorator adds each decorated function to a registry
> - new REST API returns information about all registered {{@queue_settings_calculator}} functions
> - new application_settings field for queue_settings_calculator id. UI provides drop-down to allow admin to select a queue_settings_calculator
> - Experiment Editor will call calculate_queue_settings REST API if the application interface has a queue_settings_calculator configured. Will pass the entire Experiment Model. Updates the queue settings with the returned values.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)