Improved Techniques for Caches of Search Engines Results

摘要

Result caching is an efficient technique for reducing the query processing load, hence it is commonly used in search engines. In this paper, we study query result caching and proposes a cache management policy for achieving higher hit ratios compared to traditional heuristics methods. Our cache management policy comprises an eviction policy and an admission policy, and it divides the memory allocated for caching into two parts. Our first contribution is a new set of feature-based cache eviction policies that achieve significant improvement over previous methods. Our second contribution is a study an admission policy which uses stateful features to prevent infrequent queries from polluting the cache, substantially improving the hit ratios greatly. Experiment results on two different query logs show that our policy achieves higher hit ratios when compared to previously methods.