关键词: 属性策略, 缓存替换策略, CP-ABE, 加密数据, 余弦相似度

Abstract: Abstract: Facing the needs of massive data storage , caching technology has been a kind of important means of improving the application performance of the cloud platform. The current cache replacement strategy is mainly for the plaintext data of cloud storage. In this paper, for CP-ABE encrypted data, an effective cache replacement algorithm, Minimum Attribute Value(MAV), is proposed. This algorithm combines access strategy of encrypted file , counts the number of the high-frequency attributes, and calculates the value of the file attribute based on the attribute similarity and file size. The experimental results show that the cache replacement algorithm is superior to the LRU, LFU, SIZE and other cache replacement algorithms in terms of CP-ABE encrypted data.

Key words: attribute policy, cache replacement policy, CP-ABE, encrypted data , Cosine similarity