In　this　article,　we　design　and　implement　an　air　quality　data　processing　system　based　on　big　data　technology,　which　leverages　Flume　to　acquire　data,　Kafka　to　cache　data,　and　Storm　to　process　data,　and　finally　save　the　results　to　the　in-memory　database　called　Redis　and　distributed　database　called　HBase.　In　Redis,　only　the　latest　data　of　each　monitoring　point　is　saved,　and　the　original　data　is　deleted　periodically　to　reduce　the　memory　load.　An　application　is　implemented　by　which　people　can　get　real-time　data　from　Redis　and　query　historical　data　from　HBase.　Through　experiments,　it　is　proved　that　querying　real-time　data　from　Redis　is　faster　than　relational　database.　The　system　solves　the　issues　of　poor　storage　capacity　and　slow　query　speed　when　using　relational　database　in　the　big　data　environment,　and　improves　the　efficiency　of　data　processing　significantly.　©　2018　IEEE.