In　recent　years,　with　the　rapid　development　of　Internet,　those　Query-Intensive　web　information　systems　can　not　meet　the　requirement　of　high　performance,　which　mainly　comes　from　the　following　two　aspects.　First,　most　database　systems　that　web　information　systems　use　store　the　related　data　in　disks,　which　causes　the　slow　data　access.　Second,　it　is　hard　to　support　large　numbers　of　users　accessing　in　parallel　due　to　the　limitation　of　single　server＇s　physical　resources,　which　causes　the　server　overload.　For　solving　the　problems,　this　paper　proposes　a　MMDB　(Main-Memory　Database)　cluster　based　cloud　platform　for　Query-Intensive　web　information　systems.　First,　we　design　a　MMDB　cluster　based　web　information　system　according　to　the　current　technology.　Second,　we　develop　a　model-based　cloud　platform,　which　provides　support　for　the　MMDB　cluster　based　web　information　system　at　an　architecture　level.　The　experiment　on　a　real-world　cloud　demonstrates　the　feasibility,　effectiveness　and　benefits　of　the　new　approach　to　improving　the　performance　of　those　Query-Intensive　web　information　systems.