[
https://issues.apache.org/jira/browse/CARBONDATA-4079?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Ajantha Bhat resolved CARBONDATA-4079.
--------------------------------------
Resolution: Fixed
Timestamp filter is slow when
carbon.push.rowfilters.for.vector = false
due to extra cast added in spark filtering. suggested to set this property to
true
> Queries with Date range are taking time
> ---------------------------------------
>
> Key: CARBONDATA-4079
> URL: https://issues.apache.org/jira/browse/CARBONDATA-4079
> Project: CarbonData
> Issue Type: Improvement
> Components: data-query
> Affects Versions: 2.1.0
> Reporter: suyash yadav
> Assignee: Ajantha Bhat
> Priority: Major
>
> Hi Team,
> We are doing a POC to understand how can we improve performance of the query
> fired against table created in apache carbondata.
> Below is the sample query:
>
> *spark.sql("select ts,resource,metric,value from fact_timestamp_global left
> join tags_10_Days_test on fact_timestamp_global.tags_id= tags_10_Days_test.id
> where metric in ('Outbound Utilization (percent)','Inbound Utilization
> (percent)') and resource='10.212.7.98_if:<0001>' and ts between '2020-09-21
> 00:00:00' and '2020-09-21 12:55:55' group by
> ts,resource,metric,value").show(10000,false)*
> As you can see above query contains the date range filter.We have noticed
> that due to this date range filter the query time is coming around 15 seconds
> which is not proving useful as we have to bring down the query execution time
> to 3 to 4 seconds.
> Could you please review above query and suggest a better way of framing the
> above query specially the date range filter which can be helpful to get the
> desired query execution time?
>
> In case you need more details then please do let me know.
>
> Regards
> Suyash Yadav
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
