The Attribute-weight Cache Replacement Algorithm for Encrypted Files

Abstract

For the purpose of improving the cache hit rate for encrypted files, a new cache replacement algorithm based on the access tree of files encrypted by CP-ABE - minimum similarity of attribute-weight algorithm (Minimal Attribute-Weight Similarity, MAWS) is proposed. The algorithm counts the number of attributes in the access tree and computes the weight of attribute. It creates a high frequency attribute-weight table to record the attributes that accessed by high frequency. It uses Pearson correlation coefficient and size of file to calculate the similarity of attribute-weight. The file with the minimum value of similarity of attribute-weight has the priority to be replaced. In experiment, the MAWS algorithm can get higher byte hit rate and higher hit rate of encrypted files by comparing with algorithms of FIFO (First-In First-Out), LRU (Least Recently Used), LFU (Least-Frequently-Used) and SIZE.