Member since
04-02-2018
2
Posts
0
Kudos Received
0
Solutions
04-02-2018
12:12 PM
Hi @Constantin Stanca your explanation is very helpful. I am also facing a similar problem of slow response in power BI and Microsoft team from azure has recommended us to import hive data to sql and then connect to Power BI for better performance since we are providing a live dashboard to the customers. But it is incurring us an extra cost since we have to pay for SQl transaction too . do you think that is the only option or is there a better way to provide good speed in Power BI dashboard
... View more
04-02-2018
07:39 AM
I am not able to export a transactional hive table to sqlserver. Getting the error : Store into a partition with bucket definition from Pig/Mapreduce is not supported Below is the structure of hive table create table if not exists hive_part1
(id int, name string, age int)
partitioned by (gender string)
CLUSTERED BY(id) INTO 3 BUCKETS
STORED AS ORC TBLPROPERTIES ('transactional'='true');
insert into table hive_part1 partition(gender) select * from temp_table;
--temp_table has data I am able to export orc table and partition table ( but i need to use bucketed table since i need to update the table )
This is what i am trying sqoop export --connect 'jdbc:sqlserver://XXXX;database=xxx' --username xxxx-
-password xxx --table hive_part --hcatalog-database default --hcatalog-table hive_part
--hive-partition-key gender --hive-partition-value 'male' -m 1
the above command works if the hive_part table is just orc and partitioned .
the above command works if the hive_part table is just orc and partitioned .
... View more
Labels:
- Labels:
-
Apache Hive
-
Apache Sqoop