Cache replacement algorithm [transfer]

Cache replacementAlgorithmIt is an important factor affecting the performance of the proxy cache system. A good cache replacement algorithm can produce a high hit rate. Currently, the proposed algorithms can be divided into the following three types:

(1) the traditional replacement algorithm and its direct evolution represent the following algorithms: ① LRU (least recently used) algorithm: replaces the least recently used content with the cache; ② LFU (lease frequently used) algorithm: Replace the content with the least number of visits with the cache; ③ pitkow/recker [10] proposes a replacement algorithm: if all the content in the cache is cached on the same day, replace the largest file with the cache; otherwise, replace it with the LRU algorithm.

(2) a replacement algorithm based on the key features of the cache content. Its representative algorithms include: ① size [10] replacement algorithm: replacing the largest content with the cache; ② LRU-MIN [11] replacement algorithm: This algorithm tries to minimize the number of replaced documents. Set the size of the file to be cached to S, and replace the file whose cache size is at least s according to the LRU algorithm. If no object whose size is at least S is available, the LRU algorithm is used to replace the LRU-threshold [11] algorithm with the LRU algorithm, only documents whose size exceeds a certain threshold cannot be cached. ④ lowest lacency first [12] replacement algorithm: replaces the documents with the minimum access latency with the cache.

(3) A cost-based replacement algorithm that uses a cost function to evaluate the objects in the cache. Finally, it determines the object to be replaced based on the value of substitution. The algorithms represent the following algorithms: ① hybrid [12] algorithm: The algorithm assigns a utility function to each object in the cache to replace the smallest object with the cache; ② lowest relative value [13] algorithm: replace the objects with the lowest utility value with the cache. ③ Least Normalized Cost replacement (lcnr) [14] algorithm: this algorithm uses a reasoning function about the document access frequency, transmission time, and size to determine the replacement document; ④ Bolot et al. [15] proposed a weight inference function based on the document transmission time cost, size, and last access time to determine Document Replacement; ⑤ size-adjust LRU (slru) [16] algorithm: sorts cached objects by the ratio of cost to size, and replaces objects with the smallest ratio.

In short, in order to maximize the cache hit rate, a lot of work has been carried out around the cache replacement algorithm, but the performance of the replacement algorithm depends largely on the characteristics of WWW access, there is no replacement algorithm that can outperform other algorithms in all web access modes.

Http://news.iva.cn/n469c29.shtml