Query Optimization Strategy And Implementation For Semantic Caching In Mobile Computing

Caching is a good technology to ensure the data access performance under mobilecomputing. Through the caching, we can reduce the latency and bandwidth consumption forthe mobile device to access the network. Traditional client-side caching includes page cachingand tuple caching, both of them are used navigation based on data access patterns, andsupport for associative access is very limited, in order to solve the cache based on relationaldatabase, we put forward the idea of semantic caching.Semantic caching caches the query results and the related semantic informations, use thesemantic caching to provide answers for future queries. In this paper, with the detailedanalysis of the semantic cache and the problems of research, basing on the backgroundof the mobile database applications, I applied the semantic cache mechanism to thepractical software development. And around the merging strategy and the queryoptimization for semantic caching of these two key issues, we start to study the semanticcache.With the organization of the semantic cache, this paper defines a formal logical model ofsemantic cache, and describes the organization of the semantic cache model approach. Wedivided the query into a simple query and a complex query. In view of different query, wegive a different query optimization method and different cache merging strategy.In items of semantic caching query optimization, this paper gives the classical semanticcaching query processing firstly. Furthermore, this paper proposed24rules to optimize thesemantic caching query, based on these rules, the predicate simplification algorithm forsimple queries Dis_Optimize(DP) is given and the predicate simplification algorithm forcomplex queries Conjunction_Opt(P) is given. After that, the remaining query optimizationalgorithm Remainder_Opt(Q,S) is given. The theoretical analysis proves the effectiveness ofthe optimization mechanism.In items of semantic caching merging strategy, this paper applied different kinds of querywith different merging strategy, we proposed disjunction simplification algorithm along withCompletive Merging strategy, which can get the best performance for simple queries among the alternatives. For complex queries, the dynamic merging strategy based on predicatesimplification can balance the complexity between caches and query processing and gain agood performance in a wide range.Based on the research, I’ve designed and developed an software based on Androidoperating system to query the business information, I proved the proposed queryprocessing algorithm can optimize of the availability and performance.