Abstract: queries are pervasive in massive database applications, whose execution tends to be time consuming and costly. Therefore promotion of their efficiency will largely improve the performance of the system. Semantic cache is a novel scheme for aiding query evaluation that reuses the results of previously answered queries. But little work has been done on semantic cache involving aggregate queries. This is a limiting factor in its applicability and it is mostly used in small scale database applications. In order to utilize semantic cache in massive database applications, it is necessary to extend semantic cache to support aggregate query. In this paper, query matching is identified as a foundation for answering query using semantic caches. First, a formal semantic cache model is proposed, which supports aggregate query and provides the basis for the whole research. Then the condition of query matching is presented and query matching is classified. Next, two algorithms are proposed for aggregate query matching. These two algorithms are applied to a massive database application project. Its result proves the efficiency of the algorithms.