[
https://issues.apache.org/jira/browse/MAHOUT-202?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13468626#comment-13468626
]
Sean Owen commented on MAHOUT-202:
----------------------------------
How about making a recommendation? For example in an item-based rec, you'd have
1 request to get the user's data. Say the user rated 20 items. You'd retrieve
20 items. Say each was rated by 100 people. You'd go get 2000 items' data.
That's already about 1 second.
I'm not terribly against this as it's self-contained and just adds a dependency
in a secondary module. I do just wonder whether it would be used... if you care
to use HBase you probably have well over 1M data point (or else why bother) and
it's pushing some response time issues. Compare with just dumping to a text
file and loading it into memory, where it would run much faster.
Then again this is useful perhaps if you have data in HBase already for another
reason, and, don't want to export, and, the data isn't so big. Is that the idea?
> Make Taste support HBase as data store
> --------------------------------------
>
> Key: MAHOUT-202
> URL: https://issues.apache.org/jira/browse/MAHOUT-202
> Project: Mahout
> Issue Type: New Feature
> Components: Collaborative Filtering
> Affects Versions: 0.3
> Reporter: Jeff Zhang
> Priority: Minor
> Attachments: MAHOUT-202.patch
>
>
> I'd like to add hbase as another data store option for taste.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira