[jira] [Commented] (AIRAVATA-2392) Ability to reuse previous experiment job data for a new experiment.

["Sudhakar Pamidighantam (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/AIRAVATA-2392?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16014116#comment-16014116 ] 

Sudhakar Pamidighantam commented on AIRAVATA-2392:
--------------------------------------------------

The application deployment should be able to use the reuse (indicator) and corresponding jobid to add prejob commands to fetch the right restart files for the new run using environmental variables (for local directory). 

> Ability to reuse previous experiment job data for a new experiment.
> -------------------------------------------------------------------
>
>                 Key: AIRAVATA-2392
>                 URL: https://issues.apache.org/jira/browse/AIRAVATA-2392
>             Project: Airavata
>          Issue Type: New Feature
>          Components: Airavata System, PGA PHP Web Gateway
>    Affects Versions: 0.17
>            Reporter: Eroma
>             Fix For: 0.18
>
>
> Experiments to use previous job files as inputs. User would be able to provide job ID of the previous and then use files in the working directory of that job.
> More info on the requirement:
> 1. All jobs need to run in the same cluster.
> 2. When user provides a previous job ID or experiment ID it could be even a failed job.
> 3. Users will only be abel to use their own jobs and jobs shared with them with write/read access
> 4. Application will take care of the actual file to fetch for the new job. These commands could be in the deployment or application executable in the cluster itself.



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)