topic Re: How to persist native SAS data sets to Hadoop (Hive) in SAS Data Management

How to persist native SAS data sets to Hadoop (Hive)

deepika2 — Thu, 08 Jun 2017 07:04:18 GMT

Hi,

I am having a table in Hive and using that table I am trying to create a temporary table in Hive based on some conditions and lookup values, which are stored in a SAS dataset.

I am using Execute (insert into .... ) by hadoop inside proc sql. On using the SAS dataset reference inside Execute by hadoop, it gives me an error.

Can somebody suggest how can I achieve this.

Thanks for the help.

Re: How to persist native SAS data sets to Hadoop (Hive)

LinusH — Thu, 08 Jun 2017 10:22:56 GMT

Please share the code/log so we better can understand what you are trying to do. Including the look-up step.

Re: How to persist native SAS data sets to Hadoop (Hive)

JBailey — Thu, 08 Jun 2017 13:22:38 GMT

Hi @deepika2

Anything enclosed in the parentheses is sent directly to Hive. Hive doesn't know anything about a SAS data set so you will get an error.

EXECUTE (all this stuff goes straight to Hive) BY HADOOP;

Update: I just noticed the Subject of this post. Are you trying to store SAS7BDAT files on HDFS?

Best wishes,

Jeff

Re: How to persist native SAS data sets to Hadoop (Hive)

AnnaBrown — Fri, 09 Jun 2017 20:11:36 GMT

Hi deepika2,

This article by @briankinnebrew may help you:

How to persist native SAS data sets to Hadoop (Hive)

Anna