Research on query results Cache based on log analysis in web search engines

Query results cache is an effective technology to improve the performance of Web search engines. In this paper, we conducted an analysis of Web search engine query logs from SOGOU Inc. to explore some issues of query results cache for Web search engines including the locality of Web search engine workload and the impact of cache replace policy. We also conducted a series of experiments on query results cache with a variety of cache capacity and cache replacement policy settings. Experimental results showed that when there are only a small number of historical query logs, it would be better to choose a dynamic policy, and hybrid cache approach had significant improvement than any other policies in smaller cache capacity.