An important feature
of the data preparation interface is the ability to distribute data
to a co-located data provider.
Some
of the advantages are described in Concepts in Data Preparation. The data preparation interface provides
two ways to distribute data to a co-located data provider.
When you create a prepare
data job, the
Data Output
page enables you to distribute data. For deployments that use SAS Visual Analytics Hadoop,
there is an
Add to HDFS check box and a series
of fields for specifying HDFS parameters. For deployments that use
a third-party vendor database, the check box is labeled
Add
to data server and fields are provided for specifying
the DBMS library.
For more information, see Prepare Data.If you have tables or
views that are already prepared for analysis, then you can distribute
the data without creating a prepare data job. Make sure the table
or view is registered in metadata so that it appears in the SAS Folders
tree. You can select the table object and select either
Add
to HDFS or
Add to Data Server.
The menu choice depends on the co-located data provider.
For deployments that
use a third-party vendor database as a co-located data provider, a
default library for the database is registered in metadata during
the installation and configuration time frame. This library is located
in SAS metadata at
/Products/SAS Visual Analytics Data
Preparation/SAS Visual Analytics Distributed Data.
This library is configured with options that are specific to operating
with a distributed database. For information about the options, see
SAS
Visual Analytics: Administration Guide. When you create
an add to data server job, remember to save the job so that the output
table is registered in metadata with the distributed data library.