You are viewing a plain text version of this content. The canonical link for it is here.
Posted to notifications@accumulo.apache.org by "Josh Elser (JIRA)" <ji...@apache.org> on 2015/03/06 00:01:38 UTC

[jira] [Commented] (ACCUMULO-3648) Ability to Scan Table Keys in Reverse Order

    [ https://issues.apache.org/jira/browse/ACCUMULO-3648?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14349580#comment-14349580 ] 

Josh Elser commented on ACCUMULO-3648:
--------------------------------------

This would be a very nice new feature to Accumulo. We could do it on the server at a few different levels, but I would imagine the closer we could do it to the file-level (as opposed to a Scan or Iterator), the more efficient we could make it.

For context from HBase when they implemented a reverse scanner: HBASE-4811

> Ability to Scan Table Keys in Reverse Order
> -------------------------------------------
>
>                 Key: ACCUMULO-3648
>                 URL: https://issues.apache.org/jira/browse/ACCUMULO-3648
>             Project: Accumulo
>          Issue Type: New Feature
>          Components: master, tserver
>            Reporter: Mike Fagan
>              Labels: features, performance
>
> There are cases where you would like to scan a table in reverse key order.
> For example: 
> Consider a table using an inverted timestamp key.
> This allows for very quick access to the latest records.
> But  It would be helpful to quickly access earliest records without having to:
> A) scan through all the records to build a collection to reverse sort
> B) store the table again using the reverse key



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)