[
https://issues.apache.org/jira/browse/HUDI-5909?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Hui An updated HUDI-5909:
-------------------------
Description:
For query like
{code:sql}
create table hudi_test()...;
insert into hudi_test()...;
{code}
it will create 3 hive clients(spark's hiveClient, hudi's hiveClient to create a
new hudi table, SyncTool to sync meta after the insert operation)
We can actually reuse the spark's hiveClient when creating a new hudi table.
> Reuse hive client if possible
> -----------------------------
>
> Key: HUDI-5909
> URL: https://issues.apache.org/jira/browse/HUDI-5909
> Project: Apache Hudi
> Issue Type: Improvement
> Reporter: Hui An
> Assignee: Hui An
> Priority: Major
> Labels: pull-request-available
>
> For query like
> {code:sql}
> create table hudi_test()...;
> insert into hudi_test()...;
> {code}
> it will create 3 hive clients(spark's hiveClient, hudi's hiveClient to create
> a new hudi table, SyncTool to sync meta after the insert operation)
> We can actually reuse the spark's hiveClient when creating a new hudi table.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
