閃馬智能：Serverless 如何賦能大前端？| GMTC 2021

{"type":"doc","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"2021 年 12 月 5 日至 6 日，由極客邦科技旗下的 InfoQ 中國主辦的頂級技術盛會 GMTC 全球大前端技術大會在深圳順利舉辦。GMTC 全球大前端技術大會聚焦前沿技術及實踐經驗，面向各行業對前端、移動開發、AI 技術感興趣的中高端技術人員，旨在幫助他們瞭解大前端&移動開發領域的技術趨勢與實踐案例。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"來自閃馬智能的高級研發經理\/高級架構師吳佳浩也受邀在會上發表了題爲《Serverless 賦能大前端，加速 AI 落地，閃馬的設計與思考 》的精彩演講。通過本次演講，吳佳浩向參會者介紹了 Serverless 模型，並分享了 Serverless+AI 技術的實踐和對未來技術發展的思考。"}]},{"type":"image","attrs":{"src":"https:\/\/static001.geekbang.org\/infoq\/e0\/e06dbee558a9aabe88aca297873a40d6.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"heading","attrs":{"align":null,"level":2},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"何謂 Serverless"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"Serverless，全稱 Serverless Computing（無服務器運算），又被稱爲 FaaS（Function-as-a-Service，函數即服務），是雲計算的一種模型。在 Serverless 的概念下， 計算資源以服務而非服務器的形式出現。對開發者而言，基於 Serverless 的應用部署可以在服務級別實現，而無需過多地關心服務器的問題，這就爲軟件開發、軟件迭代提供了便利。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"Serverless 並不是近幾年纔開始興起的概念。早在 2006 年，Zimki 公司便創立了第一個“按照實際調用付費”的提供服務端 JavaScript 應用的平臺（當時他們並沒有使用“Serverless”一詞，第一個使用該稱呼的是 2012 年的 Iron.io）。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https:\/\/static001.geekbang.org\/infoq\/3a\/3a2e5893441a5c662b5f8ccb2fc9ce9d.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":"center","origin":null},"content":[{"type":"text","text":"圖片來源：網絡"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"2014 年，亞馬遜推出的 AWS Lambda，引領新一輪的 Serverless 風潮，國外的 IBM、Google 等知名廠商先後跟進，而國內也有阿里雲和騰訊雲等等。但 Serverless 並不是雲廠商的專利。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"Serverless 主要解決的是在應用開發過程中的部署環境問題，以及治理和編排的架構問題。在傳統的開發過程中，所有的開發者都需要關注許多事，諸如“服務器在什麼位置？部署在什麼樣的目錄？使用何種操作系統來開發？具體用什麼語言來開發？”，還需要關注負載均衡等一系列的事情。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https:\/\/static001.geekbang.org\/infoq\/98\/983f5075dfcad7f8eff50e09b7ca9fd6.jpeg","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":"center","origin":null},"content":[{"type":"text","text":"圖片來源：閃馬智能"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"隨着時間的推移，應用開發流程中的一部分工作內容被逐一拆分出來，交由“雲端”負責統一託管，從基礎架構雲（IaaS）、容器化架構雲（CaaS）、應用平臺雲（PaaS）發展到如今的 Serverless，原本需要開發者關注的網絡、存儲、服務器、虛擬化、操作系統、中間件、運行環境、數據、應用，幾乎全部都可以在雲平臺實現。在這樣的情形下，應用開發者越來越能夠專注於“開發”本身，也就是對應用功能的開發、實現和維護了。"}]},{"type":"heading","attrs":{"align":null,"level":2},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"Serverless+AI"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"在應用 Serverless 過程中，閃馬智能內部孵化了“SMserverless”開發平臺來幫助閃馬智能的工程師們更好地完成開發工作，以應對更加複雜的需求和更緊湊的交付週期，滿足絕大多數的個性化開發場景的需求。"}]},{"type":"image","attrs":{"src":"https:\/\/static001.geekbang.org\/infoq\/72\/723aff7da50424b250ca3cb9fde797cf.gif","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https:\/\/static001.geekbang.org\/infoq\/27\/27a971b16fad94fbf0e84901af8442c2.gif","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":"center","origin":null},"content":[{"type":"text","text":"圖片來源：閃馬智能"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"如上圖所示，綠色標註的 8 個功能點，分別對應了實時視頻播放、視頻事件統計、車道流量統計、車輛行駛狀況及車輛信息展示等等功能，按照傳統的開發模式，實現這些功能大致需要一至兩個月的時間，而藉助“SMserverless”開發平臺，閃馬智能的工程師們完成這項工作僅僅需要兩天的時間。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"在“SMserverless”平臺上，開發者僅需要上傳算法模型的描述文件，即可在極短的時間內由系統響應出一個實時算法，並得到運用。目前，VisionMind 視頻智能分析平臺已經實現了通過單個或多個視頻流構建“空間”概念的功能，完成了從視頻流算法到時空數據算法的進化。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https:\/\/static001.geekbang.org\/infoq\/03\/03b4f744a3881933c78b6fd2b29ea573.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":"center","origin":null},"content":[{"type":"text","text":"圖片來源：閃馬智能"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"在“SMserverless”的幫助下，閃馬智能能做的並不止於此。目前，閃馬智能目前正在研發的第二代 ATOM 深度學習平臺，它將會爲 VisionMind 視頻智能分析平臺賦予更爲強大的認知能力。通過對“立體空間”的構建，城市道理交通系統的管理者能更加方便地獲得道路交通的運營信息，系統還將根據不同的天氣情況，對於攝像頭進行相應調整，適應場景變換。此外，在應用場景上，也將增加對於手機、平板、執法儀等終端的支持。"}]},{"type":"image","attrs":{"src":"https:\/\/static001.geekbang.org\/infoq\/4c\/4c6e795ee2c03f22dd3a3fb91b9bc6bd.gif","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":"center","origin":null},"content":[{"type":"text","text":"圖片來源：閃馬智能"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"正如吳佳浩在演講中所說，在應用開發過程中，業務場景纔是真正的“一等公民”，所有技術都是爲了業務場景去服務的，只有落到實際場景中去，技術纔是有意義的。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"目前，閃馬智能的 Vision Mind 視頻智能分析平臺和 ATOM 深度學習平臺已經在全國 200+城市落地，就應用效果而言，二者已經能夠在城市道路交通系統發揮較好的賦能效果。但這僅僅是閃馬智能的第一步，閃馬智能仍在進化。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"在未來，閃馬智能將會在尋找問題更優解的同時，嘗試將技術應用於更多的場景，更好地爲城市賦能，成爲城市空間智能管理的領導者。"}]}]}