In many implementations of FME Server the workflow consists of processing that is independent. The queue manages the incoming requests under the assumption that no two request and therefore no tow processes are dependent. This allows for the efficient allocation of jobs to the engines. This talk will discuss how FME Server was extended to manage pools of dependent processes where downstream pools are dependent on upstream pools having completed. Through the correct use of metadata, job submitters and auditing triggers the resulting extension supported interdependent process requests extremely efficiently.