The use case is, i am using prefect for my inferencing flow. from data ingestion to recieving result.
and i have multiple multiple algorithms and have assign seperate queue for each algorithm in a pool
So let say i have queue q1,q2,q3 in a work-pool inferencing connected to algorithm a1,a2 and a3.
currently theres is a priority assigned to it q1 has priority 1, q2 has priority 2 and q3 has priority 3.
so let say i need to run 50 inference in q1 and q3 and since q1 has high priority than q3 then q1 will get completed first and then q3, evene though they are independent.
Please correct if i am wrong in understanding the queue logic in a workpool.
Aim is to run inference at the same time if there is no dependency