With　the　continuous　development　of　the　Internet　of　Things　(IoT)　technology,　the　large　number　of　messages　generated　by　the　increasing　number　of　IoT　devices　has　brought　new　challenges　to　MQTT　servers.　Due　to　the　need　for　timely　forwarding　of　MQTT　messages,　the　latency　of　message　reception　and　forwarding　will　also　increase　when　there　are　too　many　clients　connected　to　the　server　and　message　publishing　and　subscription　activities　are　too　frequent.　In　response　to　this　issue,　this　paper　proposes　an　MQTT　subscription　tree　optimization　method　that　combines　hash　tables　and　trie　trees.　Then,　this　method　is　used　to　implement　the　subscription　tree　module　of　a　simple　MQTT　server,　and　its　performance　is　compared　and　analyzed　with　several　other　subscription　tree　implementation　methods.　The　final　experimental　results　show　that　when　a　large　number　of　clients　connect　to　the　server　and　frequently　publish　messages　or　subscribe　to　topics,　the　method　proposed　in　this　paper　can　reduce　message　forwarding　latency,　achieve　fast　message　routing,　and　improve　the　efficiency　of　server　message　forwarding.　©　2024　IEEE.