You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@mahout.apache.org by "Saikat Kanjilal (JIRA)" <ji...@apache.org> on 2010/11/03 05:58:25 UTC

[jira] Commented: (MAHOUT-319) SVD solvers should be gracefully stoppable/restartable

    [ https://issues.apache.org/jira/browse/MAHOUT-319?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12927734#action_12927734 ] 

Saikat Kanjilal commented on MAHOUT-319:
----------------------------------------

Hi All,
Am finally getting some time to devote to working on this task, I've started the implementation for this and have created a generic snapshotting mechanism, as part of this I am planning to make the design configurable where the user specifies two parameters to be used each time the matrix computations are run, these parameters are the storage mechanism (disk or hadoop) and the time interval on which to perform snapshotting (i.e. every 5 minutes or some similar time period).  Are there other configurable parameters that may be useful?  Also I am planning to introduce a properties file to store these parms and read them using a ResourceBundle, is there a generic properties file to use for things like this?  Looking forward for discussion on this.

> SVD solvers should be gracefully stoppable/restartable
> ------------------------------------------------------
>
>                 Key: MAHOUT-319
>                 URL: https://issues.apache.org/jira/browse/MAHOUT-319
>             Project: Mahout
>          Issue Type: Improvement
>    Affects Versions: 0.3
>            Reporter: Jake Mannix
>            Assignee: Jake Mannix
>             Fix For: 0.5
>
>
> LanczosSolver, DistributedLanczosSolver, and HebbianSolver all keep copious amounts of memory-resident data which is lost if the app crashes or is killed (OOM, forgetting to run in a screen session, and losing net connectivity to the server running it, etc...).  
> These algorithms (and many other Mahout processes!) should enable a pluggable "persist state" mechanism (to HDFS, RDBMS, local disk, key-value store, etc), and similarly, a way to pick up and start from such a state.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.