[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"tag-ampere":3},{"tag":4,"articles":10},{"id":5,"name":6,"slug":7,"article_count":8,"description_zh":9,"description_en":9},"b14988e0-beed-4dc8-a797-aab28b9922c2","Ampere","ampere",2,null,[11],{"id":12,"slug":13,"title":14,"summary":15,"category":16,"image_url":17,"cover_image":17,"language":18,"created_at":19},"68bfa04a-94c4-4c8a-921c-61e93ab207aa","cuda-cp-async-ampere-hbm-latency-en","cp.async on Ampere: Hide HBM Latency on A100","Ampere’s cp.async moves data without stalling warps, cutting HBM waits from 450–600 cycles into overlapped compute on A100.","research","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1775167612143-4qvu.png","en","2026-04-02T22:06:36.521272+00:00"]