Client-side Pooling Concepts and Overview

Prior to SAS 9.2, client-side pooling was the only option for configuring SAS Workspace servers in a pool. SAS 9.3 continues to support client-side pooling and server-side pooling. For more information about the relative strengths of each method, see these topics:

Before performing this configuration, it is important that you understand a couple of concepts. For example, what does it mean to set up a client-side pooling workspace server? If a workspace server has not been converted to pooling, then each time a SAS Web Report Studio or SAS Information Delivery Portal user starts a session, a workspace server process must be created and the user must establish a connection to this process. This can be a time-consuming sequence of events. When you set up a client-side pooling workspace server, a pool (or set) of connections to workspace servers are opened when SAS Web Report Studio or SAS Information Delivery Portal makes its first request for a workspace server. A user can then obtain a preexisting connection from the pool instead of having to establish the connection.

Another important concept is that of a puddle. A puddle is a subset of the connections in a client-side pool. Setting up puddles enables you to associate a different set of users with different puddles. The Pool Administrator is the user who reads metadata regarding the entire pool. This ID defaults to sastrust@saspw. It is not used for the actual workspace server, just to process the pool. Each puddle has a login that is specific to the puddle. Typically, the reason for setting up different groups of users is to give the different groups different levels of access.

For more information, see Choices in Workspace Server Pooling in SAS Intelligence Platform: Security Administration Guide.

This topic summarizes the steps required for setting up client-side pooling for your workspace server for use with SAS Web Report Studio and SAS Information Delivery Portal. To contrast client-side pooling with server-side pooling, see Overview of Pooling. The result of this configuration has better performance than connecting to a standard workspace server.

To set up a client-side pool, you must do the following:

If your deployment requires more than one host for client-side pooled workspace servers, then you can configure client-side pooling across multiple machines. For more information, see Configure Client-side Pooling across Multiple Machines.