falcon-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Shwetha G S (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (FALCON-26) Pipeline Monitor addition
Date Wed, 22 Jan 2014 09:41:22 GMT

    [ https://issues.apache.org/jira/browse/FALCON-26?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13878463#comment-13878463

Shwetha G S commented on FALCON-26:

1. Remove JobAction param from doSummaryJobAction() as the function is not applicable for
other actions
2. The logic of coordinatorListLe(decrementing and checking if 1) is not readable(confusing).
Instead, you should have another variable isLastCoord and set it accordingly
3. Few optimisations 
   - Instead of accumulating all coord actions into one big list and iterating later, you
can form summary in the first for loop itself. This will handle large number of instances
that span across multiple coords.
  - If the coord actions table is partitioned by time, its good to set the end time for getCoordJobInfo()
to last materialised time(as there will not be any instances beyond that).
4. Add documentation for new API

> Pipeline Monitor addition
> -------------------------
>                 Key: FALCON-26
>                 URL: https://issues.apache.org/jira/browse/FALCON-26
>             Project: Falcon
>          Issue Type: Improvement
>          Components: client
>    Affects Versions: 0.3
>            Reporter: Samarth Gupta
>            Assignee: Suhas Vasu
>         Attachments: FALCON-26-v2.patch, FALCON-26-v3.patch, FALCON-26.patch
>   Original Estimate: 120h
>  Remaining Estimate: 120h
> adding features to falcon get instance status api to return summarized result for a given
process. This will enable user to specify start and end time of instance and in return will
get summarized result grouped based on there status. and number of instance belonging to those

This message was sent by Atlassian JIRA

View raw message