
On this tab page you maintain how the event receivers are to be started.
A defined number of events is delivered by the background job in each cycle. These events can be processed sequentially or in parallel (with aRFC).
The receivers can be started synchronously (with RFC) or asynchronously (with tRFC).
Only synchronous starting of receivers is currently supported for parallel processing.
This tab page is maintained within administration of the event queue.
Features
Processing of events
Sequential
Events are delivered from the event queue in defined units (for example 5 events per read access). Sequential processing means that these events are processed in succession.
Parallel on server group
Events are delivered from the event queue in defined units (for example 5 events per read access). Parallel processing means that the events are processed at the same time in separate processes. These parallel processes can run using a defined server group.
A server group combines a number of application servers to form a logical group. If you specify a server group, the event delivery is executed using the application servers in this group.
If you do not specify a server group, the parallel execution is distributed across all application servers. You are recommended to use a server group with the appropriate resources.
Start of receivers
Synchronous
Various receivers (for example workflows) can be started for an event. Synchronous start of receivers means that they are started using a remote function call (RFC). The processing waits until the receiver call is complete.
Asynchronous
Various receivers (for example workflows) can be started for an event. Asynchronous start of receivers means that they are started using a transactional remote function call (tRFC). The processing does not wait until the receiver call is complete, but is continued immediately instead. The receivers are started asynchronously.
Sequential processing of events with synchronous start can be used for test purposes.
The delivery mode recommended in the standard version is sequential processing of events with asynchronous start of receivers.
Parallel processing (with synchronous start) can be used to distribute the system load on a predefined server group. Delivery then only uses the application servers assigned to this server group. If you do not specify a server group, the parallel execution is distributed across all application servers. You are recommended to use a server group with the appropriate resources.
You can access administration of the event queue by choosing .