[
https://issues.apache.org/jira/browse/CARBONDATA-1526?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Ravindra Pesala updated CARBONDATA-1526:
----------------------------------------
Description:
User can trigger compaction on pre-aggregate table directly, it will further merge the segments inside pre-aggregation table. To do that, use ALTER TABLE COMPACT command on the pre-aggregate table just like the main table.
For implementation, there are two kinds of implementation for compaction.
1. Mergable pre-aggregate tables: if aggregate functions are count, max, min, sum, avg, the pre-aggregate table segments can be merged directly without re-computing it.
2. Non-mergable pre-aggregate tables: if aggregate function include distinct_count, it needs to re-compute when doing compaction on pre-aggregate table.
> 10. Handle compaction in aggregation tables. Compaction of aggregation should be in syc with main table.
> --------------------------------------------------------------------------------------------------------
>
> Key: CARBONDATA-1526
> URL:
https://issues.apache.org/jira/browse/CARBONDATA-1526> Project: CarbonData
> Issue Type: Sub-task
> Reporter: Ravindra Pesala
>
> User can trigger compaction on pre-aggregate table directly, it will further merge the segments inside pre-aggregation table. To do that, use ALTER TABLE COMPACT command on the pre-aggregate table just like the main table.
> For implementation, there are two kinds of implementation for compaction.
> 1. Mergable pre-aggregate tables: if aggregate functions are count, max, min, sum, avg, the pre-aggregate table segments can be merged directly without re-computing it.
> 2. Non-mergable pre-aggregate tables: if aggregate function include distinct_count, it needs to re-compute when doing compaction on pre-aggregate table.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
Locked
