Optional Setup Tasks

It might be necessary for you to modify your initial configuration to meet specific requirements in your environment. Optional administration and configuration tasks include the following.

Task	Description
Install sas.servers as a boot script.	On UNIX machines, you can choose to install the sas.servers script as a boot script so that the SAS servers will start automatically when you start the machine. The comments in the sas.servers script contain installation instructions that are specific to your operating environment. For details, see Using the sas.servers Script on UNIX or z/OS to Start or Stop All Servers.
Optimize the performance of the metadata server.	To optimize the performance of the metadata server, you might want to adjust the maximum number of threads used by the server. See Configuring the Number of Threads Used by the Metadata Server. For other performance-related tasks, see Managing Metadata Server Memory.
Modify the configuration of your processing servers.	The following are some server configuration changes that you might want to make for performance optimization or other reasons: Define a cluster of load-balanced workspace servers in order to balance a load across workspace servers that you have installed on multiple hosts. This type of configuration is most useful when you have a large number of users (such as data integration specialists) using a workspace server for relatively long-running jobs. For details, see Understanding Server Load Balancing in SAS Intelligence Platform: Application Server Administration Guide. Modify load balancing or algorithms for stored process servers and pooled workspace servers. For details, see Understanding the Load-Balancing Algorithms in SAS Intelligence Platform: Application Server Administration Guide. If your SAS server metadata contains characters other than those typically found in the English language, then you must start your SAS server with an ENCODING= or LOCALE= system option that accommodates those characters. For details, see “Encoding and Locale Information” in the SAS Intelligence Platform: Application Server Administration Guide. If you will be submitting large jobs (for example, JAVA GRAPH jobs) on z/OS servers, then you might need to specify a larger region size. For details, see “Managing Memory” in the SAS Companion for z/OS.
Optimize Web application performance.	For information about configuring middle-tier components for better efficiency and performance, see Best Practices for Middle-Tier Configuration in SAS Intelligence Platform: Middle-Tier Administration Guide.
Adjust server logging.	You can change your server configurations to obtain additional logging information for troubleshooting, performance management, or other purposes. For details, see Administering Logging for SAS Servers and SAS Logging: Configuration and Programming Reference.
Enable job and report scheduling.	To enable SAS scheduling features, you can install and configure scheduling software, configure operating system scheduling, or configure in-process scheduling. For details, see Scheduling in SAS.
Increase Java heap memory allocation for desktop applications.	To process large amounts of data, you might need to modify the Java heap memory allocation that is configured for SAS Management Console, SAS Data Integration Studio, SAS OLAP Cube Studio, and SAS Information Map Studio. see Managing the Java Heap Used By Desktop Applications in SAS Intelligence Platform: Desktop Application Administration Guide.
Set up change management for SAS Data Integration Studio users.	If you want to implement change management for SAS Data Integration Studio, then you must set up project repositories and at least one change-managed folder. For details, see Administering SAS Data Integration Studio in SAS Intelligence Platform: Desktop Application Administration Guide.
Collect ARM log information for SAS Data Integration Studio batch jobs.	If you want to collect ARM log information for SAS Data Integration Studio jobs that are run on a batch basis, then you must enable logging for the batch server that executes the jobs. For details, see Collecting ARM Log Information for SAS Data Integration Studio Jobs.