LR-LRU: A PACS-Oriented Intelligent Cache Replacement Policy

Abstract

An intelligent cache replacement policy suitable for picture archiving and communication systems (PACS) was proposed in this work. By combining the logistic regression (LR) algorithm with the classic least recently used (LRU) cache replacement policy, we have created a new intelligent cache replacement policy called LR-LRU. The LR-LRU policy is unlike conventional cache replacement policies, which are solely dependent on the intrinsic properties of the cached items. Our PACS-oriented LRLRU algorithm identifies the variables that affect file access probabilities by mining medical data. The LR algorithm is then used to model the future access probabilities of the cached items, thus improving cache performance. In addition, l1-regularization was used to reduce the absolute values of the variables' coefficients. This screens some variables that have little influence on the model by causing their coefficients to approach zero, which achieves the effect of screening the variables. Finally, a simulation experiment was performed using the trace-driven simulation method. It was shown that the l1-regularized LR model is superior to the LR and l2-regularized LR models. The LR-LRU cache algorithm significantly improves PACS cache performance when compared to conventional cache replacement policies, such as LRU, LFU, SIZE, GDF, and GDSF.