Answer: Hash join
Explanation: Pipeline parallelism is a way in which multiple path depend on one another.But there is a possibility of overlapping of execution and the output of one step is released as input to consecutive step. Hash join is used for the joining of the algorithm applied on equal joins and is separated in form of columns and so does not follow the overlap in the execution and is much slower join for the pipeline parallelism .