Hi Kumar,

You can create a Hive table with Parquet format, and then let upstream
application output the parquet files to this Hive table's HDFS location
(each time can be in a new partition). Then Kylin will be able to fetch the
new data incrementally from this table.

Hive provides the logic layer over the physical model, which is
recommended.


2017-11-14 14:30 GMT+08:00 Kumar, Manoj H <[email protected]>:

> Pls. advise on this – Input source of data – Can it from Parquet File
> format instead of Hive table? Upstream system generating the File format as
> output. Pls. suggest. This will be important features to have it.
>
>
>
> Manoj
>
> This message is confidential and subject to terms at: http://
> www.jpmorgan.com/emaildisclaimer including on confidentiality, legal
> privilege, viruses and monitoring of electronic messages. If you are not
> the intended recipient, please delete this message and notify the sender
> immediately. Any unauthorized use is strictly prohibited.
>



-- 
Best regards,

Shaofeng Shi 史少锋

Reply via email to