Discovery and Load Balancing APIs

An IBM® Communications Server for Windows application program developer can locate services and load balance among those services using the TCP/IP protocol. There are three basic methods that an application program can take advantage of this new function:
Method 1:
Communications Server SNA APIs (LUx (RUI/SLI), APPC, CPIC). Using the APIs will get the support basically for free if an existing application is already written to an SNA API. With this method, no new code must be written to take advantage of the location/load balancing functions. The only constraint with this method is that API code expects the client's configuration data to reside in an INI file, or LDAP Communications Server for Windows.
Method 2:
Service Location Protocol (SLP) User Agent (UA) API. An SLP UA DLL is packaged with the product which provides support for Communications Server service location and load-balancing over TCP/IP connections. This method provides the greatest flexibility for the application developer in terms of how to do the service location/load-balancing, where to obtain client configuration, and how to present these functions to the end user.
Method 3:
Using a combination of UA (for location) and the QEL/MU CM_CSLIST_GETII primitive for load-balancing (for 3270 and LU 6.2 applications only). This method is a hybrid of the first two in that it reduces the amount of code needed to be written to only the location function and gives maximum flexibility in terms of client configuration.

IBM recommends using the API client for location and load-balancing. If the application developer is unable to do so or desires to support Telnet, method 2 is provided. If support for QEL/MU is already provided then method 3 may be used. Since the first method is really nothing new from the application developer's perspective, the following discussion applies to the last two methods.