As a novel computing paradigm, the concept of edge computing has emerged in recent years. Compared with traditional cloud computing, edge computing can improve the performance of many applications. In edge computing, data storage services can be provided by edge servers, which are usually closer to the data sources for better Quality of Service. Data storage-related issues in edge computing are important and should be carefully addressed. In this paper, data allocation and replication are investigated in edge computing. Each data source needs to upload and replicate its data either to the Cloud, or to the edge server. Three important goals related to the data uploading and replication locations are explored, including maximizing total data access frequencies supported by the edge server, maximizing the total replicas, and maximizing the number of data sources which upload their data to the edge server. The tradeoff among these goals and different solutions are presented as well. Performance evaluation demonstrates that proposed solutions can achieve these goals well, respectively.