人工智能正在接管DevOps嗎？

{"type":"doc","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"對於許多開發人員來說，預測DevOps的下一步已經成爲他們最喜歡的一種消遣。在過去的十年裏，我們看到我們的行業迅速變化，在同一時期，程序員的角色也發生了根本性的變化。"}]},{"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":"事實上，尤其是那些（我們應該說）特定年齡的開發者中，“傳統”開發者的角色似乎已經不復存在。與傳統的軟件開發生命週期（即軟件的規劃、構建和發佈）不同，我們中的許多人現在都在多職責團隊中工作，在這些團隊中，開發和運維並存，並且越來越難以區分。"}]},{"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":"這種模式被稱爲"},{"type":"link","attrs":{"href":"https:\/\/en.wikipedia.org\/wiki\/DevOps","title":"","type":null},"content":[{"type":"text","text":"DevOps"}]},{"type":"text","text":"，但只是在最近開始流行。而且DevOps本身只花了幾年（或者幾個月？）就開始改變和適應。現在，我們正漂泊在首字母縮略詞的海洋中（包括最近的"},{"type":"link","attrs":{"href":"https:\/\/www.devsecops.org\/","title":"","type":null},"content":[{"type":"text","text":"DevSecOps"}]},{"type":"text","text":"），這些首字母縮略詞表示一種稍微不同的工作方式，每個首字母縮略詞對未來的看法也略有不同。"}]},{"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":"其中最新的是"},{"type":"link","attrs":{"href":"https:\/\/en.wikipedia.org\/wiki\/Artificial_Intelligence_for_IT_Operations","title":"","type":null},"content":[{"type":"text","text":"AIOps"}]},{"type":"text","text":"。在這個願景中，AI工具"},{"type":"link","attrs":{"href":"https:\/\/www.infoq.com\/news\/2021\/07\/AI-IT-operations\/","title":"","type":null},"content":[{"type":"text","text":"正在慢慢取代開發者的角色"}]},{"type":"text","text":"——就像DevOps以前做的那樣——並最終將完全取代DevOps。"}]},{"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":"評估這一預測是否正確是一件棘手的事情，但在本文中，我們仍將嘗試。我們將看看AI承諾爲開發過程做些什麼，評估它是否真的能夠從人類開發者手中接管，然後看看DevOps在幾十年後可能會是什麼樣子。"}]},{"type":"heading","attrs":{"align":null,"level":2},"content":[{"type":"text","text":"自動化的前景"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"首先，爲了理解爲什麼AI工具被DevOps團隊如此迅速地採用，瞭解它們承諾能夠實現什麼是非常有用的。這一承諾基本上可以分爲兩部分："}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"bulletedlist","content":[{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"一方面，AI工具的使用“僅僅”是過度緊張的開發人員掌握他們所使用的日益複雜的系統的一種方式。"}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"另一發面，AI工具提供了一系列自動化代碼開發和部署技術，從根本上改變了軟件的製作方式。"}]}]}]},{"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":"關於前者——讓我們的生活更輕鬆的AI工具——我們只需看看過去十年中"},{"type":"link","attrs":{"href":"https:\/\/www.infoq.com\/articles\/devops-and-cloud-trends-2021\/","title":"","type":null},"content":[{"type":"text","text":"雲基礎設施模型的激增"}]},{"type":"text","text":"，就知道爲什麼人工智能是必要的。如今，大多數公司都使用某種形式的混合或雲基礎設施，如果沒有某種形式的人工智能幫助，管理這些基礎設施幾乎是不可能的。事實上，混合和多種雲基礎設施、容器和超大規模應用程序等微服務架構，創造了一個比以往任何時候都"},{"type":"link","attrs":{"href":"https:\/\/www.esg-global.com\/blog\/whats-in-your-hybrid-cloud-infrastructure","title":"","type":null},"content":[{"type":"text","text":"更復雜"}]},{"type":"text","text":"的企業IT環境。"}]},{"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":"其次是人工智能工具採用的第二個方面——事實上，一些人工智能工具，如GitHub的人工智能編碼助手或微軟的DeepDev，不僅讓我們的生活更輕鬆，還爲代碼開發和分發開闢了新的可能性。例如，人工智能系統評估單個用戶需求的速度，使得使用AI來調整和"},{"type":"link","attrs":{"href":"https:\/\/bitadvisors.com\/education\/yahoo-search-virus-mac-safari-chrome","title":"","type":null},"content":[{"type":"text","text":"指定每個用戶的默認Web瀏覽行爲"}]},{"type":"text","text":"成爲可能，從而降低了它們在使用我們的軟件時對特定類型惡意軟件的脆弱性。"}]},{"type":"heading","attrs":{"align":null,"level":2},"content":[{"type":"text","text":"DevOps、AIOps和NoOps"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"由於AI工具的實用性，除了最頑固的DevOps團隊，它們已經被所有團隊廣泛迅速採用。事實上，對於現在運行多個不同雲的團隊（差不多是所有團隊，非常多），人工智能接口幾乎已經成爲他們"},{"type":"link","attrs":{"href":"https:\/\/www.infoq.com\/articles\/evolve-and-scale-devops\/","title":"","type":null},"content":[{"type":"text","text":"發展和擴展他們的DevOps程序"}]},{"type":"text","text":"的必要條件。"}]},{"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":"這種轉變最明顯、最具體的結果是開發人員花時間查看的數據和系統。例如，過去運維團隊的一個主要角色是構建和維護一個儀表盤，所有工作人員都可以查看這個儀表盤，該儀表盤包含一個軟件上的所有適當數據。"}]},{"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":"如今，這一中心任務已經基本過時。隨着軟件變得越來越複雜，用一個儀表盤包含某一特定軟件的所有相關信息的想法開始聽起來很荒謬。相反，大多數DevOps團隊在使用人工智能工具“自動”監控他們正在使用的軟件，並且只有在明確出現問題時才顯示數據。"}]},{"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":"這是我們作爲開發人員和運維人員工作方式的重大轉變，因此它被賦予了自己的首字母縮略詞——AIOps，也就不足爲奇了。事實上，有些人甚至更進一步，聲稱這種對人工智能工具的依賴現在意味着我們"},{"type":"link","attrs":{"href":"https:\/\/searchitoperations.techtarget.com\/definition\/NoOps","title":"","type":null},"content":[{"type":"text","text":"正在進入NoOps時代"}]},{"type":"text","text":"。然而，關於哪個角色被取消存在分歧——無論NoOps是指“沒有開發人員”還是“沒有運維”。"}]},{"type":"heading","attrs":{"align":null,"level":2},"content":[{"type":"text","text":"AI革命"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"所有這些關於軟件開發哲學背景的思考聽起來可能有點抽象。直到你看到數字，從中可以明顯看出，人工智能已經對軟件的發佈速度（和質量）產生了巨大的影響。"}]},{"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":"GitLab最近對4000多名開發人員進行了調查，得出了一些確切的數據。調查發現，一些公司發佈新代碼的速度比以前快了十倍。很明顯，"},{"type":"link","attrs":{"href":"https:\/\/about.gitlab.com\/developer-survey\/","title":"","type":null},"content":[{"type":"text","text":"75%的用戶使用人工智能和機器學習"}]},{"type":"text","text":"來測試和審查預發佈的代碼。這比一年前的40%略高。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https:\/\/imgopt.infoq.com\/fit-in\/1200x2400\/filters:quality(80)\/filters:no_upscale()\/articles\/ai-devops-takeover\/en\/resources\/2pasted%20image%200-1633456731453.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":"paragraph","attrs":{"indent":0,"number":0,"align":"center","origin":null},"content":[{"type":"link","attrs":{"href":"https:\/\/www.altexsoft.com\/blog\/engineering\/software-testing-qa-best-practices\/","title":"","type":null},"content":[{"type":"text","text":"圖片來源"}]}]},{"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":"這對於開發人員來說是個好消息，或者至少對於那些希望快速生成大量代碼的人來說是個好消息。不幸的是，這並不總是與生成優秀代碼相同。例如，從勒索軟件攻擊趨勢可以明顯看出，測試不良的代碼正迅速"},{"type":"link","attrs":{"href":"https:\/\/www.atlantic.net\/hipaa-compliant-hosting\/the-growing-threat-of-ransomware-in-2021\/","title":"","type":null},"content":[{"type":"text","text":"成爲許多組織的一個顯著漏洞源"}]},{"type":"text","text":"，人工智能驅動的測試系統的出現對減少這一漏洞幾乎沒有什麼作用。"}]},{"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":"儘管如此，前進的方向是顯而易見的。再過幾年，DevOps團隊中的絕大多數似乎都將依賴人工智能工具，軟件的發佈速度將是以前的數倍。這讓我們回到了我們開始的問題——考慮到人工智能工具現在做了這麼多DevOps工作，我們還需要DevOps人員嗎？"}]},{"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":"好吧，既是，也不是。原因如下。"}]},{"type":"heading","attrs":{"align":null,"level":2},"content":[{"type":"text","text":"挑戰"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"在最膚淺的層面上，可以通過一個非常簡單的測試來回答這個問題。隨着人工智能工具在整個開發部門的應用，開發人員的工作量是否減少了？問一個普通開發者的問題，你會得到一個笑聲。答案是否定的。"}]},{"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":"這是因爲，隨着DevOps團隊在其軟件的日常管理上花費的時間越來越少，他們過去在這方面花費的時間現在被更具價值的任務佔用了——戰略規劃、元分析，以及確保他們的開發目標與管理目標一致。事實上，許多人"},{"type":"link","attrs":{"href":"https:\/\/devops.com\/why-devops-will-cease-to-exist\/","title":"","type":null},"content":[{"type":"text","text":"認爲會使DevOps過時"}]},{"type":"text","text":"的人工智能“革命”似乎使團隊比以往任何時候都更龐大、更繁忙。這是因爲它的副作用是使開發週期大大加快。"}]},{"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":"這一點得到了業務管理者的充分理解。在最近"},{"type":"link","attrs":{"href":"https:\/\/www.zdnet.com\/article\/artificial-intelligence-taking-over-devops-functions-survey-confirms\/","title":"","type":null},"content":[{"type":"text","text":"對ZDNet的採訪"}]},{"type":"text","text":"中，iTech AG的執行副總裁Matthew Tiani指出，DevOps現在“通過增強的技術工具集（包括源碼管理、CI\/CD、編排）得到了增強。”"}]},{"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":"他還補充說，DevOps的成功實施利用了“一種兼容的開發方法，例如敏捷和scrum，以及一種促進和鼓勵開發人員和運維人員之間協作的組織承諾。”"}]},{"type":"heading","attrs":{"align":null,"level":2},"content":[{"type":"text","text":"學會共存"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"實際上，這些因素和趨勢意味着DevOps團隊越來越關注業務目標，而不是技術挑戰。這當然是一個變化，但這對於我們生產的軟件的質量應該不是一個消極的變化。可以說，事實上，人工智能工具使團隊能夠將人力資源集中在他們最擅長的地方——創造性、整體性和戰略性任務中。"}]},{"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":"你不需要找太遠的證據。舉個例子，Vue JS現在是世界上增長最快的開發框架，現在已經"},{"type":"link","attrs":{"href":"https:\/\/bestwebhostingaustralia.org\/internet-statistics\/","title":"","type":null},"content":[{"type":"text","text":"有240000多個在線網站"}]},{"type":"text","text":"是用這個框架構建的。在人工智能工具出現之前，一個新的複雜的開發框架可以如此迅速地在全世界推廣的想法是可笑的。現在，在人工智能工具的支持下，DevOps團隊有了更多的自由來勇敢地做出決策，相信他們的工具足夠先進來應對技術挑戰。"}]},{"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":"這意味着人工智能工具不太可能取代DevOps，或者至少在短期內不會取代DevOps。正如我們"},{"type":"link","attrs":{"href":"https:\/\/www.infoq.com\/articles\/devops-secure-trends\/","title":"","type":null},"content":[{"type":"text","text":"在採用DevOps的之前趨勢中"}]},{"type":"text","text":"看到的那樣，開發團隊仍然需要戰略領導，無論他們的技術工具如何先進。而且，儘管人工智能現在在許多領域都有很強的能力，但它們仍然缺少以真正創造性的方式響應用戶需求的能力。"}]},{"type":"heading","attrs":{"align":null,"level":2},"content":[{"type":"text","text":"最後"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"總而言之，我們可能還"},{"type":"link","attrs":{"href":"https:\/\/www.infoq.com\/news\/2021\/05\/developers-e2e-tests\/","title":"","type":null},"content":[{"type":"text","text":"需要人力開發人員"}]},{"type":"text","text":"一段時間。儘管他們的角色可能與十年前完全不同，但這就是在如此發展變化的行業工作的代價（和樂趣）。"}]},{"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","marks":[{"type":"strong"}],"text":"作者介紹"},{"type":"text","text":"："}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"link","attrs":{"href":"https:\/\/www.linkedin.com\/in\/sambocetta","title":"","type":null},"content":[{"type":"text","text":"Sam Bocetta"}]},{"type":"text","text":" 曾是一名安全分析師，他職業生涯的大部分時間在海軍擔任網絡工程師。他現在已經半退休，向公衆傳授安全和隱私技術。Sam的大部分工作涉及彈道系統的滲透測試。他分析了我們的網絡，尋找切入點，然後根據我的發現創建了安全漏洞評估。此外，他還幫助規劃、管理和執行復雜的黑客攻擊演練，以識別漏洞並降低海軍（路上和海上）使用的企業系統的風險態勢。他的大部分工作集中在識別和防止應用程序威脅和網絡威脅，降低攻擊向量區域，消除漏洞和常規報告。他能夠發現弱點並制定新的戰略，支持我們的網絡抵禦一系列網絡威脅。Sam與架構師和開發人員密切合作，爲應用程序中發現的漏洞確定緩解控制措施，並執行安全評估，以模擬各種威脅的策略、技術和程序。"}]},{"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","marks":[{"type":"strong"}],"text":"原文鏈接："}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"link","attrs":{"href":"https:\/\/www.infoq.com\/articles\/ai-devops-takeover\/","title":"","type":null},"content":[{"type":"text","text":"Is Artificial Intelligence Taking over DevOps?"}]}]}]}