The invention provides a big data self-adaption topology processing method based on a storm.The method comprises the steps of user request monitoring, request information analyzing, topology logic generating and data flow query.Operation operator are selected according to different needs of different users by monitoring the customized service request of the users in real time, services are dynamically customized, topology logic of storm recognizable operation is generated, and therefore dynamic customizing and real-time processing of the data processing services are achieved; the corresponding operators are conveniently selected through an operator execution tree, the optimal operation execution topology meeting the customizing requests of the users is formed, the pertinence of data flow query and mining is enhanced, and the timeliness of data is guaranteed as much as possible.