A Novel Approach to Cache Replacement Policy Model Based on Genetic Algorithms

Effective buffering of disk blocks in primary memory as one of the considerable factors for enhancing the performance any system. The block replacement policy is very significant design choice that specify that disk block must be substituted while completion of the buffer is done. This paper introduced a suggested self-tuning cache replacement policy which forecasts future request of a block based on its access pattern. It has an online responsive behavior to changing access patterns by executing genetic algorithms (GAs). The proposed policy constantly balances among the potential attributes of the object in cache memory: recency, frequency, size, and time. The suggested approach is tested very well and assessed.