Understanding SAS/CONNECT Servers

The Uses of SAS/CONNECT in the SAS Intelligence Platform

Overview of the Uses of SAS/CONNECT in the SAS Intelligence Platform

Introduction to SAS/CONNECT explains the general capabilities that SAS/CONNECT gives SAS programmers. This section explains how some of the SAS Intelligence Platform clients and some of the SAS macros that are provided with the platform use SAS/CONNECT. You must install SAS/CONNECT on the correct machines (and possibly configure a SAS/CONNECT spawner and server) to use the following features:

the ability of SAS Data Integration Studio and SAS Enterprise Miner to generate SAS/CONNECT code
the ability of SAS Data Integration Studio and SAS Enterprise Miner to use grid computing
the ability to promote the contents of a metadata repository from one environment to another (for example, from a development environment to a test environment)

For more information about providing the infrastructure for these features, see the following sections.

SAS/CONNECT, SAS Data Integration Studio, and SAS Enterprise Miner

If you have installed and configured SAS/CONNECT, then both SAS Data Integration Studio and SAS Enterprise Miner can generate SAS/CONNECT code as part of a job. For example, suppose that you are creating a SAS Data Integration Studio job that processes a large amount of remote data. Instead of sending requests for data across the network and results being returned across the network to you, the job can use SAS/CONNECT compute services to submit the code that does the processing to a SAS/CONNECT server that is running on the remote machine, as shown in the following figure.

Using a SAS/CONNECT Server to Access Remote Data

[Using a SAS/CONNECT Server to Access Remote Data]

Because the code is now executing on the machine where the data resides, the job can execute much faster.

Alternatively, the job can use the data transfer service in SAS/CONNECT to download the data from the remote host so that the job can run on the local processor. If necessary, the job can also upload revised data.

For complete details about how to set up your environment in order to enable this functionality, see Setting Up Multi-Tier Environments in SAS Intelligence Platform: Desktop Application Administration Guide. For information about creating data integration and data mining jobs, see the user's guides and online Help for SAS Data Integration Studio and SAS Enterprise Miner.

SAS/CONNECT and Grid Computing

If you have invested in the SAS Grid Manager software application, then users of SAS Data Integration Studio and SAS Enterprise Miner can use SAS/CONNECT not only to submit code to a remote host, but to submit code to a grid of processors. The following figure illustrates the role that SAS/CONNECT plays in grid computing.

The Role of SAS/CONNECT in Grid Computing

[The Role of SAS/CONNECT in Grid Computing]

The illustration is designed to give you an idea of how this type of system is set up. For more information, see What is SAS Grid Computing? in Grid Computing in SAS.

SAS/CONNECT and Metadata Replication

SAS/CONNECT plays an important role in the area of metadata replication. The Metadata Manager plug-in to SAS Management Console provides a Replication Wizard to copy an entire metadata repository from one environment to another. The wizard produces jobs that use SAS macros, and the SAS macros use SAS/CONNECT code to copy the repository data sets between environments.

The job that is submitted to the workspace server in the source environment (the development environment in this case) uses PROC UPLOAD in SAS/CONNECT to copy the source repository data sets to the target environment. For more information, see Using the Replication Wizard to Copy an Entire Metadata Repository in SAS Intelligence Platform: System Administration Guide.

Top of Page