Aimed　at　the　potential　overload　problems　caused　by　the　uncertainty　of　the　arrived　data　in　a　data　stream　management　system,　the　load　shedding　issue　of　the　data　stream　system　is　researched.　Focus　on　the　condition　that　exists　join　operators　of　two　streams,　a　semantic-based　load　shedding　technique　is　applied.　A　data　stream　load　shedding　model　is　designed　and　implemented,　which　solves　four　problems　including　load　shedding　and　anti-shedding　time,　amount,　location　and　selecting　predicate.　The　experiment　result　is　analyzed,　which　shows　that　the　load　shedding　model　presented　can　effectively　avoid　the　low　processing　efficiency　when　system　is　overload,　and　guarantee　the　coordination　of　arrived　data　and　system　processing　capability,　and　make　a　combination　of　system　stability　and　correctness　as　a　result.　©　2008　IEEE.