As　the　most　existing　stream　clustering　algorithms　can　not　generate　online　clustering　results　in　real-time,　an　online　data　stream　clustering　algorithm　is　proposed　by　using　sliding　windows　and　density-based　grid　storage　structure.　The　algorithm　achieves　a　rapid　speed　for　online　clustering　data　stream　and　it　can　provide　users　with　real-time　clustering　results　and　reflect　the　dynamic　evolution　of　data　streams.　Experimental　results　show　that　the　algorithm　proposed　has　a　good　capacity　of　dealing　with　rapid　evolutional　data　stream　and　have　a　good　clustering　quality.