A performance analysis of semantic caching for distributed semi-structured query processing

Abstract

Caching is important for any system attempting to achieve high performance. The semantic caching is an approach trying to benefit from certain semantical knowledge of the data to be processed. The expectation is that semantical information might help to reduce the number of cache misses and in certain cases even avoid queries to the primary data. However, the major obstacle for wide application of semantic caching is the query containment problem which is computationally hard. In this paper we introduce an approximate conservative algorithm for semantic caching of semistructured queries and analyze its applicability for distributed query processing. Based on this analysis, we outline few scenarios where semantic caching can be benefitial for query processing in a distributed system of heterogeneous semi-structured information resources. © 2010 Springer-Verlag.