The cross-system coupling (XCF) services provide the following
functions that a multisystem application or subsystem programmer can
use:
- A way to define a collection of unique parts of a program, and
a way for each part to identify the other parts so they can work together.
- A way for program parts to send messages to or receive messages
from other parts on the same MVS™ system
or on a different one, without regard for the I/O considerations involved.
Messages can be sent without knowing specifically where the receiving
part resides.
- A way to monitor the program parts that you (the programmer) define
to XCF. XCF maintains information about the parts you define, and
provides notification of changes. Again, these parts can be on the
same MVS system or different MVS systems.
- A way to design your program for high availability, such that
primary parts are on one system and backup parts are on another system.
When the primary system fails, XCF notifies the backup parts on the
other system and the backup parts can be designed to take over the
function of the primary. The primary and backup parts can also be
running in different address spaces on the same system. In this case,
the parts running in the backup address space can be designed to take
over when the primary address space fails.
- A way to allow batch jobs and started tasks to be restarted automatically.
You can use the XCF recovery function, automatic restart management,
to design your application for high availability by allowing it to
be restarted automatically when it, or the system it is running on,
fails. See Using the Automatic Restart Management Function of XCF for more information.
Examples of exploiting XCF appear in various topics, as the XCF services
are explained. Before learning more about the XCF services and how
to use them, you must understand some basic XCF concepts.