In this topic, you will deploy a data flow that was created
in the Design Studio using SQL warehousing operators. You will first
convert the data flow into an XML file. Then import (deploy) the XML
into the DataStage server.
About this topicShown below is a data flow created
in the Design Studio using SQL warehousing operators.
You want to deploy
this data flow onto the DataStage Server that is installed on a Windows
computer. Save this data flow as dfexpression. Validate the
data flow and then perform the following steps:
Procedure
- With the data flow displayed in the canvas, click anywhere
inside the Design Studio canvas.
Note: Make sure that none
of the operators and connectors in the data flow are selected.
- Click the Properties tab. In the SQL
Execution Database list, select CVSAMPLE as
the database schema.
- On the Data Flow menu, click Deploy
as Data Stage Executables. A directory dfexpression_datastage is
created inside the generated-code directory of your data warehouse
project. This directory contains a metadata XML file, a parallel job
file and an SQL script file as shown below.
- Click the DataStage Servers view. Right-click the DataStage
server where you want to deploy the XML metadata file and select Deploy
job metadata to server.... The Import DataStage
Project Elements window opens.
- In the File Name text field, click
the ellipsis icon (...) and select the XML
metadata file dfexpression.xml to be deployed.
- Set up the location of the DataStage client. Click . Click the ellipsis
(...) icon and browse to the directory where
the DataStage client is installed.
- Click Import to selected project and
then select the DataStage project where you want to deploy the metadata
file.
- Click Import. The Import
statistics are displayed to you. Click Finish to
close the Import DataStage Project Elements window and return to the
Design Studio explorer.
- If the DataStage server is installed on a machine different
from the one where the Design Studio is installed, you must perform
the following additional steps:
- Copy the sqw_runsqlscript.ksh file from the Design
Studio machine to the DataStage Server. This script file
is located inside the \Client\workspace\eclipse\plugins\com.ibm.datatools.sqwruntime.tasks\scripts directory
of the InfoSphere™ Warehouse installation
directory.
- Copy the generated SQL file dataflowname.sql from
the Design Studio machine to the DataStage Server. This
SQL file is located inside the \Client\workspace\data
warehouse project name\generated-code\data flow
name_datastage directory of the InfoSphere Warehouse installation
directory.
- Inside the Job Status view, right-click the job and select Launch
DataStage Designer. The job sequence corresponding to
the data flow opens in the DataStage Designer as shown below.
Note: If
the job name is not displayed inside the Job Status view, you might
have to close the Design Studio instance and restart the Design Studio
again. Refresh the Job Status view. The job name must appear in the
Job Status view.
- Inside the DataStage Designer, double-click the Execute_Command_0,
and browse to the full path on the DataStage server where the sqw_runsqlscript.ksh file
is saved. Save the job and close the job.
- Return to the Design Studio, right-click the job in the
Job Status view and click Compile Job. The job status changes to compiled.
Note: If the job name is not displayed inside the Job Status view, you might have to close the Design Studio instance and restart the Design Studio again. Refresh the Job Status view. The job name must appear in the Job Status view.