Compute-intensive　and　latency-sensitive　applications　place　stringent　demands　on　mobile　devices　in　terms　of　computational　power　and　task　latency.　Edge　computing　is　a　promising　technique　to　alleviate　the　computational　constraints　of　mobile　terminals　and　reduce　their　energy　consumption　through　computation　offloading.　This　paper　proposes　a　two-tier　task　offloading　framework　for　multi-user　multi-server　edge　IoT　scenarios,　comprehensively　considering　factors　such　as　task　offloading　decision,　network　channel　allocation,　edge　server　computational　resources,　device　transmission　power,　and　the　size　of　the　task　offloading　data　volume,　and　constructing　a　multi-user　task　offloading　Optimization　problem　model.　In　the　process　of　solving　the　optimization　problem,　a　Combining　GA　and　PSO　Computation　offloading　Algorithm　(CGPCA)　is　proposed　for　solving　the　model.　Finally,　the　convergence　of　the　algorithm　is　verified　through　simulation　experiments,　and　the　experimental　results　show　that　the　proposed　algorithm　in　this　study　can　effectively　reduce　the　task　offloading　delay　and　energy　consumption,　and　improve　the　quality　of　end-user　service　compared　with　other　existing　algorithms.　©　2024　IEEE.