hive-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Prasanth Jayachandran (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (HIVE-20621) GetOperationStatus called in resultset.next causing incremental slowness
Date Sat, 22 Sep 2018 00:59:00 GMT

     [ https://issues.apache.org/jira/browse/HIVE-20621?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Prasanth Jayachandran updated HIVE-20621:
-----------------------------------------
    Description: 
Fetching result set for a result cache hit query gets slower as more rows are fetched. For
fetching 10 row result set it took about 900ms but fetching 200 row result set took 8 seconds. 

Reason for this slowness is GetOperationStatus is invoked inside resultset.next() and it happens
for every row even after operation has completed. This is one RPC call per row fetched (there
is also connection overhead without keepalive). 

  was:
Fetching result set for a result cache hit query gets slower as more rows are fetched. For
fetching 10 row result set it took about 900ms but fetching 200 row result set took 8 seconds. 

Reason for this slowness is GetOperationStatus is invoked inside resultset.next() and it happens
for every row even after operation has completed. This is one RPC call per row fetched. 


> GetOperationStatus called in resultset.next causing incremental slowness
> ------------------------------------------------------------------------
>
>                 Key: HIVE-20621
>                 URL: https://issues.apache.org/jira/browse/HIVE-20621
>             Project: Hive
>          Issue Type: Bug
>          Components: JDBC
>    Affects Versions: 4.0.0, 3.2.0
>            Reporter: Prasanth Jayachandran
>            Assignee: Prasanth Jayachandran
>            Priority: Major
>         Attachments: HIVE-20621.1.patch
>
>
> Fetching result set for a result cache hit query gets slower as more rows are fetched.
For fetching 10 row result set it took about 900ms but fetching 200 row result set took 8
seconds. 
> Reason for this slowness is GetOperationStatus is invoked inside resultset.next() and
it happens for every row even after operation has completed. This is one RPC call per row
fetched (there is also connection overhead without keepalive). 



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Mime
View raw message