Benefits of using OTMA

IMS applications that do not use OTMA do not need to be modified to use OTMA, unless they use SETO calls, in which case, you might need to modify them to use OTMA.

The SETO call applies to APPC/IMS and SPOOL/API processing.

For each OTMA-originated transaction, the SETO call returns a status code. You can bracket the SETO call with an INQY call if necessary.

Full-duplex processing provides an environment in which transactions and output messages are sent and processed in parallel.

You can implement IMS device support outside IMS. You can also implement device support for your IMS subsystem that is different from what IMS provides, or enable device support that IMS does not provide. The following figure illustrates how IMS can communicate with a device, shown here as a workstation, using device support implemented within an OTMA client. IMS device support using Virtual Telecommunications Access Method (VTAM®) is shown for comparison.
Figure 1. IMS communicates with a device using device support implemented within an OTMA client
IMS communicates through XCF and device support to communicate with device (workstation). Alternately, device support through VTAM enables communication with device (workstation).

Flow-control and transaction-processing attributes are dynamically bound to the transaction.

Clients have high-performance access to IMS:
  • OTMA uses the z/OS® cross-system coupling facility application programming interface (API).
  • OTMA does not use VTAM and IMS device-dependent support.

Transactions based on different protocols (that is, that have different processing requirements such as being recoverable or irrecoverable) can be associated with a single transaction pipe.

You can connect up to 255 clients to the OTMA group.

Messages can be extended using the user-data section of the message prefix, allowing additional user information to be sent with the transaction.

User information and transaction pipe name are included within the messages themselves.

Different clients can specify the same transaction pipe names, instead of needing to use uniquely named resources.

You do not need to use networking architectures, such as SNA (Systems Network Architecture).