BPM 11g and Human Workflow Shadow Rows by Adam Desjardin
By Juergenkress-Oracle on Jun 07, 2014
During the OFM Forum last week, there were a few discussions around the relationship between the Human Workflow (WF_TASK*) tables in the SOA_INFRA schema and BPMN processes. It is important to know how these are related because it can have a performance impact. We have seen this performance issue several times when BPMN processes are used to model high volume system integrations without knowing all of the implications of using BPMN in this pattern.
Most people assume that BPMN instances and their related data are stored in the CUBE_*, DLV_*, and AUDIT_* tables in the same way that BPEL instances are stored, with additional data in the BPM_* tables as well. The group of tables that is not usually considered though is the WF* tables that are used for Human Workflow. The WFTASK table is used by all BPMN processes in order to support features such as process level comments and attachments, whether those features are currently used in the process or not.
For a standard human task that is created from a BPMN process, the following data is stored in the WFTASK table:
- One row per human task that is created
- The COMPONENTTYPE = "Workflow"
- TASKDEFINITIONID = Human Task ID (partition/CompositeName!Version/TaskName)
- ACCESSKEY = NULL
For regular information on Oracle SOA Suite become a member in the SOA & BPM Partner Community for registration please visit www.oracle.com/goto/emea/soa (OPN account required) If you need support with your account please contact the Oracle Partner Business Center.