A FEPI application runs in a front-end CICS® system and accesses applications in a back-end CICS or IMS™ system. FEPI lets it do this by simulating a terminal connected to the back-end system; this means that it has to act just like a real terminal and terminal operator.
The back-end systems are known as targets and the connections to them are arranged in pools that define the properties controlling communication. Targets, pools, and properties are defined by your system programmer, who can tell you which targets and pools to use and what properties they have.
When a connection has been established, on successful completion of a bind, the connection is in session and it can be allocated by FEPI for a conversation with the back-end system.
Conversations are the basis of all FEPI applications and, depending upon the needs of your application, may be used in several ways (see FEPI application design):
The task that started the conversation owns it and other tasks cannot issue commands for it; however, the owning task can transfer ownership to another task. You can have as many conversations as you like at a time with various targets: they can be consecutive or, much more usefully, interleaved.
FEPI simulates a 3270-type terminal (SLU2 mode) for both CICS and IMS systems; it also supports the SLU P mode that is used by IMS for programmable terminals such as the 4700 family. The mode to be used, SLU2 or SLU P, is a property of the pool being used. Your application cannot change the mode of a conversation.
The data that you send and receive can be formatted or data stream and, as with mode, the data type is a property of the pool being used:
The same basic set of FEPI commands is used for all modes and data types and protocols, but the command options and keywords are generally different.
[[ Contents Previous Page | Next Page Index ]]