Setting Up Multi-Tier Environments

Access Remote Data

Overview of Remote Data Access

The procedures in this section explain how to enable your SAS Data Integration Studio users to access data on a host that is remote from the SAS Workspace Server. This is done by installing SAS/CONNECT software on the SAS Workspace Server host and then configuring a SAS/CONNECT server on the remote host.
In order to access data on a host that is remote from your SAS Workspace Server, you must run SAS/CONNECT software. Run SAS/CONNECT on your SAS Workspace Server machine (the local host), and on the machine with your data (the remote host). To understand this setup, consider the following figure:
three network machines
In this network, we have three machines:
  • the machine running the SAS Workspace Server and the SAS Metadata Server (the local host)
  • a separate machine with data (the remote host)
  • the SAS Data Integration Studio client machine

Set Up a Local Host

These procedures assume that SAS software has been installed on the local host machine. Furthermore, the local host machine needs to be set up as follows:
  • The local host runs the SAS Workspace Server.
  • The SAS Application Server of which the SAS Workspace Server is a component needs to be set as the default SAS Application Server for SAS Data Integration Studio. This is done from the SAS Data Integration Studio client machine. See Connect to a Workspace Server .
  • SAS/CONNECT software needs to be installed on the local host machine. This should have been done at installation time.

Set Up a Remote Host

Install and configure Base SAS and SAS/CONNECT software on the remote host. Follow the instructions in the SAS Intelligence Platform: Installation and Configuration Guide.
The installation process creates the server initialization script ConnectServer.bat. Use this script to invoke the SAS/CONNECT spawner as a service. The spawner starts SAS/CONNECT sessions as needed. You can find the files in the SAS-installation-directory\Lev1\appServer\ConnectServer.
You can make other configuration changes to the remote host from any machine running SAS Management Console.
In the next two sections, you define the SAS/CONNECT server and your SAS library in metadata.

Set Up a Remote SAS/CONNECT Server

On the remote host, use the SAS Deployment Wizard to define a second and new SAS Application Server that contains a SAS/CONNECT server and a SAS Workspace Server. The new servers are necessary because the local SAS Application Server does not have access to the remote library. For information about using the SAS Deployment Wizard to create new a SAS Application Server, see the SAS Intelligence Platform: Application Server Administration Guide.

Define SAS Libraries

From SAS Management Console, define one or more SAS libraries on the remote host machine. The libraries contain the tables that are accessed by SAS Data Integration Studio. These instructions are appropriate for SAS data sets:
  1. Expand Data Library Manager.
  2. Right-click Libraries and select New Library.
  3. Use the New Library Wizard to define the library. Assign the new library to the new SAS Application Server on the remote host.

Set Up a SAS Data Integration Studio Client Machine

From the SAS Data Integration Studio client machine, you need to do the following:
  1. define a default SAS Application Server
  2. define tables in the metadata for the remote host
  3. test your setup
In addition, if you have created a new authentication domain for your remote server, you need to create new logins for your SAS Data Integration Studio users. These logins must contain a user ID and password that are valid on the SAS/CONNECT host.

Register Source Tables on the Remote Host

To identify in metadata the tables on the remote host, follow these steps:
  1. In SAS Data Integration Studio, select Filethen selectRegister Tables.
  2. In the Sources folder, select a source, such as SAS, and click Next.
  3. In the SAS Library box, select the library you created earlier.
  4. Verify or edit the library details.
  5. Consider selecting the DBMS check boxes to enable case-sensitivity or special characters for the source tables, and then click Next.
  6. Select the tables to register in metadata and click Next.
  7. Review your entries, and then click Finish.

Test Your Setup Using View Data

To test your setup, follow these steps:
  1. In SAS Data Integration Studio, select the Inventory tab.
  2. In the Inventory tab, expand the Library node.
  3. Right-click a table in the library and select View Data.
This test verifies that you can access your remote library.

Process Remote Jobs

Overview of Remote Job Processing

You might want to process one or more SAS Data Integration Studio jobs. There might be a large amount of data on a remote machine to which you save your results. In such an instance, it might be more efficient to move the job itself to the remote machine. A multi-tiered configuration can make that possible.
nultitier diagram
To move a job to a remote machine, you need to have a SAS Workspace Server on the remote host. The following software must be installed on the remote host machine:
  • Base SAS software
  • SAS Integration Technologies software
If you have not already installed this software, follow the instructions in the SAS Intelligence Platform: Installation and Configuration Guide. After installation, run the SAS Deployment Wizard to configure the following:
  • SAS object spawner
  • SAS Workspace Server
  • SAS DATA step batch server

Select Your Application Server and Workspace Server

To run SAS Data Integration Studio jobs on a remote host, open SAS Data Integration Studio and select Toolsthen selectOptionsthen selectSAS Server. Select the remote SAS Application Server, and then click OK.

Register Your Batch Job Deployment Directory

Batch jobs run through the Platform Scheduler, as configured through the Schedule Manager.
To register the directory for your batch jobs, follow these steps:
  1. Create the directory on the specified host.
  2. In SAS Management Console, expand the Schedule Manager and select Deployment Directories.
  3. With the name of the new SAS Application Server selected, select New.
  4. In the New Directory window, enter a directory name and path, and then click OK.

Deploy a Job Remotely Using SAS Data Integration Studio

To test your remote job setup, follow these steps:
  1. Start SAS Data Integration Studio.
  2. In the Inventory tab, expand Jobs.
  3. Right-click the job you that want to deploy and select Schedulingthen selectDeploy.
  4. In the Deploy a job for scheduling window, select a batch server and change the name, deployment directory, and job location, as necessary.
  5. Click OK to deploy the job.
The deployed code is automatically copied to the remote host and is ready for scheduling or running at any time.
Copyright © SAS Institute Inc. All rights reserved.