About Uppi

Uppi · ‎08-01-2016

Hi, We tried it and performance was pretty much same , difference of couple of seconds. But if we can get official confirmation, we can surely move ahead without any worries for production environment. I see Vectorization support for String only as a major push for going with String datatype. Regards

Uppi · ‎08-01-2016

Hi, Can you please confirm the datatype we should use for VARCHAR. Will String be OK? Can I confirm that STRING and VARCHAR performance will be same? Regards

Uppi · ‎07-29-2016

Hi, We have prototyping a datawarehouse solution on Hadoop/Hive. One of the issue arised was whther we should use same datatype e.g. varchar2 in Oracle converts to varchar in Hive. number in Oracle to int in Hive, date in Oracle to timestamp in Hive etc. I want to understand whether any performance issue arise if we use STRING for VARCHAR2, BIGINT for NUMBER? This helps us to fasten the DDL process if we have fixed number of datatypes without loss of data/precisions Regards Upendra

Online	Offline
Last Visited	‎05-24-2016 04:37 PM

Member Since	‎05-23-2016 07:31 AM
Last Visited	‎05-24-2016 04:37 PM
Posts	4
Kudos received	2

Cloudera Community

Re: Hive STRING vs VARCHAR Performance

Re: Hive STRING vs VARCHAR Performance

Hive STRING vs VARCHAR Performance