Edge collaborative replica placement method

The invention relates to an edge collaborative replica placement method, and belongs to the field of edge computing. The method comprises three parts: dividing an edge cooperation area based on the similarity between base station contents and base station users; the edge resource manager predicts copy popularity in the region and completes copy placement recommendation; and the cloud center node optimizes a copy deployment rule of each edge cooperation domain through a reinforcement learning algorithm so as to solve local redundancy of copies in the edge nodes and improve user service quality.