Data block activeness valuation and data migration method based on access behavior

The invention relates to a data block activeness valuation and data migration method based on access behaviors, and belongs to the technical field of data storage. According to the method, the hottest data block and the coldest data block are screened out through heat statistics for subsequent migration. In order to establish an accurate model, the method selects a proper statistical period, a proper preheating period and a proper time window to perform access statistics according to business characteristics, the hottest data block and the coldest data block in each storage layer are found out, and the heat of each data block is obtained by counting the read-write access frequency and IO characteristics of the data block. Meanwhile, heat statistical data of each hour is regularly weighted and accumulated along with time. The weight is reduced as time goes on, that is, the more new the heat statistical value period is, the higher the weight is. According to the statistical period configured by the user, the we