[
https://issues.apache.org/jira/browse/HBASE-2605?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12870456#action_12870456
]
Andrew Purtell commented on HBASE-2605:
---------------------------------------
Yeah a trillion is deliberately crazy but might make the test prohibitively
expensive to run. A billion would be fine.
> full system application test scenario: sensor network mining simulation
> -----------------------------------------------------------------------
>
> Key: HBASE-2605
> URL: https://issues.apache.org/jira/browse/HBASE-2605
> Project: Hadoop HBase
> Issue Type: Sub-task
> Reporter: Andrew Purtell
>
> Try to insert a trillion rows of simulated time series data as floating point
> values as byte[]. Run scans while writing and over the final result. Compute
> aggregates. Verify counts and metrics reported by the writers to within some
> acceptable threshold.
> Consider a mix of online updates and incremental bulk import (HBASE-1923).
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
