Hi,
If you are using 7.x, you can spawn multiple child processes and then create a sync engine to join back to parent using correlation. The sync engine will publish correlation doc to parent only when all child processes are completed. Also it depends how do you want to save data (either use task data or custom database).
My suggesstion is not to depends on task data as its not permanent (you will have to cleanup / archive that), application data should be in the custom database.
In 8.x, i think you can spawn multiple process in a single step.
Regards,
Sumit
#webMethods#webMethods-BPMS#BPM