Parallel processing
Splitting one run's work across several work processes so that it finishes sooner, and collecting what each part returned before the run goes on. Async processingAsync processingDoing work outside the request that triggered it, with a guarantee that it eventually happens — deferred jobs, retries, and periodic work. Distinct from messaging, which is about telling other systems.SAP and community, narrow by design is the neighbouring concern: work deferred past the request with a guarantee that it eventually happens, where parallel processing is throughput inside a run that waits for its results.
Overview
CALL FUNCTION STARTING NEW TASK with DESTINATION IN GROUP runs remote-enabled function modules in parallel on a group of application servers, and RECEIVE RESULTS reads back each call's output. In ABAP for Cloud Development the released class CL_ABAP_PARALLEL runs instances of a class that implements IF_ABAP_PARALLEL.
Not covered: At the single call. On the classic path each unit of work is a remote-enabled function module with its own task ID and callback, and the calling program decides how the work is split and collects the results itself.
How the core statements and the community wrappers came to be present side by side has not been assessed. SAP's SPTA framework is named in the zconcurrency_api README and was not checked on an SAP page.
Wrappers that take a class, a report or a method instead of a function module and hand the results back to the caller, running over aRFC, SPTA or batch jobs, several of them shaped after Java threads or JavaScript promises.
Artifacts
Content
Details
Coverage
For your agent
- Ref
capability:parallel-processing- JSON
- /capabilities/parallel-processing.json
- MCP
- Connecting an agent