[jira] [Resolved] (IMPALA-1405) Performance Improvements for HBase Scan

["Tim Armstrong (Jira)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/IMPALA-1405?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Tim Armstrong resolved IMPALA-1405.
-----------------------------------
    Resolution: Later

This is not specific enough to keep open for years...

> Performance Improvements for HBase Scan
> ---------------------------------------
>
>                 Key: IMPALA-1405
>                 URL: https://issues.apache.org/jira/browse/IMPALA-1405
>             Project: IMPALA
>          Issue Type: New Feature
>          Components: Perf Investigation
>    Affects Versions: Product Backlog
>            Reporter: Martin Grund
>            Priority: Minor
>              Labels: performance
>
> This is a ticket to track the efforts to improve the performance of the HBase scan by first batching intermediate results and reducing the number of JNI calls between backend and the HBase client.
> The plan is to provide a layer between the backend and the HBase client that takes care of building the HBase scan and buffering of intermediate results. Optimally this layer uses allocated memory from the backend to directly store the tuples in Impalas native format.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)