[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"article-ergo-hestia-pricing-time-to-market-databricks-zh":3,"article-related-ergo-hestia-pricing-time-to-market-databricks-zh":33,"series-industry-4a2fbd38-b5c2-4590-9d4b-87f39f95ab9c":76},{"id":4,"slug":5,"title":6,"content":7,"summary":8,"source":9,"source_url":10,"author":11,"image_url":12,"cover_image":12,"category":13,"language":14,"translated_content":11,"related_article_id":15,"keywords":16,"key_takeaways":25,"views":29,"created_at":30,"published_at":31,"topic_cluster_id":32},"4a2fbd38-b5c2-4590-9d4b-87f39f95ab9c","ergo-hestia-pricing-time-to-market-databricks-zh","ERGO Hestia 4 招縮短定價上線","\u003Cp data-speakable=\"summary\">這篇在講 ERGO Hestia 如何用 Databricks 把定價資料、模型\u003Ca href=\"\u002Fnews\u002Fopenai-service-terms-app-risk-users-zh\">服務\u003C\u002Fa>和治理整合起來，縮短上線時間。\u003C\u002Fp>\u003Cp>看完這 5 個做法，你可以判斷自己該先改資料流、模型呼叫、治理控管，還是部署方式，避免像原系統那樣在 100+ 模型、1,000+ 變數下出現 10x 到 20x 的延遲尖峰。\u003C\u002Fp>\u003Ctable>\u003Cthead>\u003Ctr>\u003Cth>項目\u003C\u002Fth>\u003Cth>規格 A\u003C\u002Fth>\u003Cth>規格 B\u003C\u002Fth>\u003Cth>規格 C\u003C\u002Fth>\u003C\u002Ftr>\u003C\u002Fthead>\u003Ctbody>\u003Ctr>\u003Ctd>Lakebase\u003C\u002Ftd>\u003Ctd>Delta tables 上的線上特徵層\u003C\u002Ftd>\u003Ctd>免抽取作業\u003C\u002Ftd>\u003Ctd>持續同步\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>\u003Ca href=\"https:\u002F\u002Fwww.databricks.com\u002Fproduct\u002Fmosaic-ai\">Mosaic AI Model Serving\u003C\u002Fa>\u003C\u002Ftd>\u003Ctd>直接 API 存取模型\u003C\u002Ftd>\u003Ctd>毫秒級請求路徑\u003C\u002Ftd>\u003Ctd>少一層 adapter\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>\u003Ca href=\"https:\u002F\u002Fwww.databricks.com\u002Fproduct\u002Funity-catalog\">Unity Catalog\u003C\u002Fa>\u003C\u002Ftd>\u003Ctd>資料與模型統一治理\u003C\u002Ftd>\u003Ctd>血緣可追蹤\u003C\u002Ftd>\u003Ctd>版本與保留一致管理\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>Azure DevOps CI\u002FCD\u003C\u002Ftd>\u003Ctd>分階段部署\u003C\u002Ftd>\u003Ctd>降低切換風險\u003C\u002Ftd>\u003Ctd>保留既有 ETL\u003C\u002Ftd>\u003C\u002Ftr>\u003C\u002Ftbody>\u003C\u002Ftable>\u003Ch2>1. Lakebase 把線上特徵留在 lakehouse\u003C\u002Fh2>\u003Cp>Lakebase 讓 ERGO Hestia 把定價所需的線上特徵，直接放在 \u003Ca href=\"https:\u002F\u002Fwww.databricks.com\u002Fproduct\u002Flakebase\">Lakebase\u003C\u002Fa> 與 Delta tables 的組合裡，不必再匯出到 PostgreSQL。這一步先解決的是資料搬運本身，而不是先去優化模型。\u003C\u002Fp>\n\u003Cfigure class=\"my-6\">\u003Cimg src=\"https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1781697768906-9krk.png\" alt=\"ERGO Hestia 4 招縮短定價上線\" class=\"rounded-xl w-full\" loading=\"lazy\" \u002F>\u003C\u002Ffigure>\n\u003Cp>對定價系統來說，少一次抽取，就少一次延遲與失真來源。Sync Tables 讓處理後資料和服務資料持續對齊，團隊也不用靠人工批次補資料。\u003C\u002Fp>\u003Cul>\u003Cli>資料留在 Databricks 內部\u003C\u002Fli>\u003Cli>Sync Tables 自動更新\u003C\u002Fli>\u003Cli>不再依賴外部資料庫匯出\u003C\u002Fli>\u003C\u002Ful>\u003Ch2>2. Mosaic AI Model Serving 直接接模型\u003C\u002Fh2>\u003Cp>\u003Ca href=\"https:\u002F\u002Fwww.databricks.com\u002Fproduct\u002Fmosaic-ai\">Mosaic AI Model Serving\u003C\u002Fa> 取代了原本夾在定價引擎和服務資料庫之間的 adapter 層。請求現在直接打到託管端點，路徑更短，維護點也更少。\u003C\u002Fp>\u003Cp>這種改法最適合需要毫秒級回應的場景。少掉自建快取和轉接邏輯後，白天流量波動時比較不容易被某一層卡住。\u003C\u002Fp>\u003Ccode>Pricing engine -> Model Serving Endpoint -> response in milliseconds\u003C\u002Fcode>\u003Ch2>3. Unity Catalog 讓模型和資料共用治理面\u003C\u002Fh2>\u003Cp>\u003Ca href=\"https:\u002F\u002Fwww.databricks.com\u002Fproduct\u002Funity-catalog\">Unity Catalog\u003C\u002Fa> 把資料、模型版本、血緣和保留政策放進同一個控制面。對保險業這種受監管環境來說，這不只是管理方便，而是能回答「這個價格\u003Ca href=\"\u002Fnews\u002Fmanaged-chatgpt-access-policy-layers-zh\">怎麼\u003C\u002Fa>算出來」的必要條件。\u003C\u002Fp>\n\u003Cfigure class=\"my-6\">\u003Cimg src=\"https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1781697768992-d0eq.png\" alt=\"ERGO Hestia 4 招縮短定價上線\" class=\"rounded-xl w-full\" loading=\"lazy\" \u002F>\u003C\u002Ffigure>\n\u003Cp>它也讓測試更安全。定價專家可以在同一個受治理環境裡驗證 live data、做 A\u002FB 測試，還能保留完整歷史紀錄，方便事後稽核。\u003C\u002Fp>\u003Cul>\u003Cli>模型版本先註冊再服務\u003C\u002Fli>\u003Cli>訓練資料保留可追溯\u003C\u002Fli>\u003Cli>資料層與模型層都有血緣\u003C\u002Fli>\u003C\u002Ful>\u003Ch2>4. 分階段遷移，比一次切換更穩\u003C\u002Fh2>\u003Cp>ERGO Hestia 沒有一次把整套系統翻掉，而是保留既有 ETL，先把同步資料導到 Lakebase，再逐步擴大使用範圍。這種做法降低風險，也讓每一步都能先驗證。\u003C\u002Fp>\u003Cp>如果你的業務不能停機，這會比大改寫更實際。它的重點不是加更多工具，而是把原本多餘的跳轉一層一層拿掉，讓新舊系統可以並行。\u003C\u002Fp>\u003Cul>\u003Cli>既有 ETL 繼續運作\u003C\u002Fli>\u003Cli>同步路徑從 PostgreSQL 轉到 Lakebase\u003C\u002Fli>\u003Cli>\u003Ca href=\"https:\u002F\u002Fazure.microsoft.com\u002Fproducts\u002Fdevops\">Azure DevOps\u003C\u002Fa> 負責部署流程\u003C\u002Fli>\u003C\u002Ful>\u003Ch2>5. 先改資料流，再改定價速度\u003C\u002Fh2>\u003Cp>最後的結果，是一套能隨市場變化即時反應的定價架構，不必再等排程刷新。對 ERGO Hestia 來說，這代表定價模型可以更快上線，也能支援 real-time B2C pricing。\u003C\u002Fp>\u003Cp>對一個跑 100+ 模型、1,000+ 變數的團隊，真正的價值不只是快，而是把定價\u003Ca href=\"\u002Fnews\u002Fopencode-terminal-ai-coding-loop-zh\">變成\u003C\u002Fa>可持續迭代的成長機制，從資料進來到對外決策都更清楚。\u003C\u002Fp>\u003Ch2>怎麼挑\u003C\u002Fh2>\u003Cp>如果你最痛的是資料搬運，先看 Lakebase；如果卡在模型呼叫延遲，先改 Mosaic AI Model Serving；如果你最怕稽核和版本混亂，Unity Catalog 應該先上。\u003C\u002Fp>\u003Cp>若你正在現有生產系統上做現代化，且不能接受高風險重寫，最值得抄的是分階段遷移。ERGO Hestia 的案例說明，提速通常不是靠多加一層，而是把不必要的 hop 拿掉。\u003C\u002Fp>","4 個架構調整讓 ERGO Hestia 把定價上線時間縮短，並在單一 lakehouse 內治理 100+ 模型。","www.databricks.com","https:\u002F\u002Fwww.databricks.com\u002Fblog\u002Fhow-ergo-hestia-reduced-time-market-lakebase-and-mosaic-ai-model-serving",null,"https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1781697768906-9krk.png","industry","zh","83a3e653-a35b-4a2e-9f92-d2db22d4deb6",[17,18,19,20,21,22,23,24],"ERGO Hestia","Databricks","Lakebase","Mosaic AI Model Serving","Unity Catalog","Azure DevOps","定價系統","lakehouse",[26,27,28],"把資料、模型服務和治理放進同一個 lakehouse，能直接縮短定價上線時間。","Lakebase、Mosaic AI Model Serving、Unity Catalog 和分階段遷移，分別對應資料流、模型呼叫、治理與部署風險。","對 100+ 模型、1,000+ 變數的系統來說，減少 hop 比增加工具更能穩定降低延遲。",0,"2026-06-17T12:02:22.440161+00:00","2026-06-17T12:02:22.427+00:00","caa87b65-9bbc-46fe-bba8-4f4158dd2d8b",{"tags":34,"relatedLang":35,"relatedPosts":39},[],{"id":15,"slug":36,"title":37,"language":38},"ergo-hestia-pricing-time-to-market-databricks-en","ERGO Hestia cut pricing time-to-market with Databricks","en",[40,46,52,58,64,70],{"id":41,"slug":42,"title":43,"cover_image":44,"image_url":44,"created_at":45,"category":13},"9e33a02c-97e7-4646-9db7-09fc3ee4bd5a","2-billion-nvidia-coherent-ai-plant-huang-warning-zh","2億美元 Nvidia-Coherent AI 廠擴建，黃仁勳再提 AI 規則…","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1781704974176-q77e.png","2026-06-17T14:02:28.980686+00:00",{"id":47,"slug":48,"title":49,"cover_image":50,"image_url":50,"created_at":51,"category":13},"1f259ae1-4769-4de4-980e-429b719bb889","huang-marvell-ai-thesis-hyperscale-infrastructure-zh","黃仁勳一句話，把 Marvell 從題材變論點","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1781703214058-ndp7.png","2026-06-17T13:33:05.013659+00:00",{"id":53,"slug":54,"title":55,"cover_image":56,"image_url":56,"created_at":57,"category":13},"2a151488-09f9-4aa8-a654-3f1d9d7e159c","china-ai-open-source-efficiency-global-sales-zh","中國 AI 轉向：開源、效率、出海","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1781702271555-i3q3.png","2026-06-17T13:17:25.59471+00:00",{"id":59,"slug":60,"title":61,"cover_image":62,"image_url":62,"created_at":63,"category":13},"0cf56d85-887b-4fb1-8589-046da6513d26","openai-oracle-universal-credits-enterprise-buying-zh","OpenAI 進 Oracle 企業採購圈","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1781696892976-sx90.png","2026-06-17T11:47:35.092555+00:00",{"id":65,"slug":66,"title":67,"cover_image":68,"image_url":68,"created_at":69,"category":13},"dd3d240a-0f53-49a4-90a5-cac17171f3fd","managed-chatgpt-access-policy-layers-zh","4 層規範決定企業版 ChatGPT 可怎麼用","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1781695973066-pbtw.png","2026-06-17T11:32:17.633521+00:00",{"id":71,"slug":72,"title":73,"cover_image":74,"image_url":74,"created_at":75,"category":13},"c826a181-b373-4a9e-a494-1f8f4bc86c3c","openai-service-terms-app-risk-users-zh","OpenAI 服務條款把第三方 App 風險留給使用者","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1781695063951-v71m.png","2026-06-17T11:17:21.223004+00:00",[77,82,87,92,97,102,107,112,117,122],{"id":78,"slug":79,"title":80,"created_at":81},"ee073da7-28b3-4752-a319-5a501459fb87","ai-in-2026-what-actually-matters-now-zh","2026 AI 真正重要的事","2026-03-26T07:09:12.008134+00:00",{"id":83,"slug":84,"title":85,"created_at":86},"83bd1795-8548-44c9-9a7e-de50a0923f71","trump-ai-framework-power-speech-state-preemption-zh","川普 AI 框架瞄準電力、言論與州權","2026-03-26T07:12:18.695466+00:00",{"id":88,"slug":89,"title":90,"created_at":91},"ea6be18b-c903-4e54-97b7-5f7447a612e0","nvidia-gtc-2026-big-ai-announcements-zh","NVIDIA GTC 2026 重點拆解","2026-03-26T07:14:26.62638+00:00",{"id":93,"slug":94,"title":95,"created_at":96},"4bcec76f-4c36-4daa-909f-54cd702f7c93","claude-users-spreading-out-and-getting-better-zh","Claude 用戶更分散，也更會用","2026-03-26T07:22:52.325888+00:00",{"id":98,"slug":99,"title":100,"created_at":101},"bd903b15-2473-4178-9789-b7557816e535","openclaw-raises-hard-question-for-ai-models-zh","OpenClaw 逼問 AI 模型價值","2026-03-26T07:24:54.707486+00:00",{"id":103,"slug":104,"title":105,"created_at":106},"eeac6b9e-ad9d-4831-8eec-8bba3f9bca6a","gap-google-gemini-checkout-fashion-search-zh","Gap 把結帳搬進 Gemini","2026-03-26T07:28:23.937768+00:00",{"id":108,"slug":109,"title":110,"created_at":111},"0740e53f-605d-4d57-8601-c10beb126f3c","google-pushes-gemini-transition-to-march-2026-zh","Google 把 Gemini 轉換延到 2026 年 3…","2026-03-26T07:30:12.825269+00:00",{"id":113,"slug":114,"title":115,"created_at":116},"e660d801-2421-4529-8fa9-86b82b066990","metas-llama-4-benchmark-scandal-gets-worse-zh","Meta Llama 4 分數風波又擴大","2026-03-26T07:34:21.156421+00:00",{"id":118,"slug":119,"title":120,"created_at":121},"183f9e7c-e143-40bb-a6d5-67ba84a3a8bc","accenture-mistral-ai-sovereign-enterprise-deal-zh","Accenture 攜手 Mistral AI 賣主權 AI","2026-03-26T07:38:14.818906+00:00",{"id":123,"slug":124,"title":125,"created_at":126},"191d9b1b-768a-478c-978c-dd7431a38149","mistral-ai-faces-its-hardest-year-yet-zh","Mistral AI 迎來最硬的一年","2026-03-26T07:40:23.716374+00:00"]