降低網絡AI應用開發門檻，AIOps兩大部署模式來助力

{"type":"doc","content":[{"type":"blockquote","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"​​​​​​​​​​​​​​​​摘要：企業運維需求及挑戰，來看看華爲AIOps如何解決！","attrs":{}}]}],"attrs":{}},{"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":"本文分享自華爲雲社區","attrs":{}},{"type":"link","attrs":{"href":"https://bbs.huaweicloud.com/blogs/280543?utm_source=infoq&utm_medium=bbs-ex&utm_campaign=other&utm_content=content","title":"","type":null},"content":[{"type":"text","text":"《【雲駐共創】AIOps？企業運維新力量！》","attrs":{}}]},{"type":"text","text":"，原文作者：啓明。","attrs":{}}]},{"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":"國際慣例，我們先介紹一下AIOps的概念：AIOps，即 ArtificialIntelligence for IT Operations，智能運維，將人工智能應用於運維領域，基於已有的運維數據（日誌、監控信息、應用信息等），通過機器學習的方式來進一步解決自動化運維沒辦法解決的問題。","attrs":{}}]},{"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":"Gartner預測，當前的IT應用程序會發生劇變，而且管理整個IT生態系統的方式也會改變。這些變化的關鍵是Gartner所稱的AIOps平臺。","attrs":{}}]},{"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":"我們今天要討論的，就是AIOps的需求挑戰，以及我們通過怎麼樣的方式去應對這種挑戰。","attrs":{}}]},{"type":"heading","attrs":{"align":null,"level":1},"content":[{"type":"text","text":"AIOps需求及挑戰","attrs":{}}]},{"type":"heading","attrs":{"align":null,"level":2},"content":[{"type":"text","text":"（一）新技術、新挑戰，呼喚高度智能的電信網絡","attrs":{}}]},{"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":"近年來，以5G爲代表的新技術在電信網絡中得到了快速的應用。新技術的應用，給我們帶來了很多的收益，比如大連接、低時延、高速率等等。5G的發展，讓這些數據都至少有一個數量級的提升。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/f7/f7506f204a92d2b763937064e350667d.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":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"​但是，數據量級的提升，伴隨着的，是運維難度的增加，從而給運維帶來了如下挑戰：","attrs":{}}]},{"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","attrs":{}}],"text":"1. 網絡複雜性：","attrs":{}}]},{"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","attrs":{}}],"text":"數據量級的增大，讓網絡變得更加複雜","attrs":{}},{"type":"text","text":"：新技術得到了快速應用，舊技術卻沒有同步退出，導致我們每引入一項新技術，都需要在原來的複雜度上做一個加法。而在某些場景式，甚至要去做乘法。","attrs":{}}]},{"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":"比如，在無線領域，2G/3G/4G/5G，“四代同堂”；在覈心網，PS/CS/MS物聯網等等十域並存......如此高的網絡複雜度勢必會給運維帶來相當大的挑戰。","attrs":{}}]},{"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","attrs":{}}],"text":"2. 2B新需求","attrs":{}}]},{"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":"運維的第二個挑戰是To B的新場景，也就是企業應用。5G的應用推動了智能製造，網絡也逐步融入到了企業的生產製造流程當中。在這種情況下，對網絡可靠性的要求必然會提高，畢竟網絡一旦出問題，生產流程就可能會受影響，甚至會中斷，這樣造成的損失將會非常大。","attrs":{}}]},{"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","attrs":{}}],"text":"3. 成本壓力","attrs":{}}]},{"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":"成本壓力主要是由前面兩個挑戰傳導而來。前兩個挑戰導致我們要麼面臨一個比較複雜的網絡，要麼就是有更高的要求。如果我們以傳統的運維方式去應對的話，必然會導致成本的急劇上升。當然，成本的提高，還有一個因素就是能耗。畢竟，5G的能耗要遠高於4G的能耗。","attrs":{}}]},{"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":"針對上述這些挑戰，我們要如何去應對呢？","attrs":{}},{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"AI技術是關鍵。","attrs":{}}]},{"type":"heading","attrs":{"align":null,"level":2},"content":[{"type":"text","text":"（二）AI是提升電信網絡自動化和智能化的關鍵技術","attrs":{}}]},{"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":"在運維成本方面，有統計顯示，90%的運維都需要人工去參與，而70%的成本就是人力成本。在這種情況下，一個很自然的想法就是能不能使用AI的技術來降低人的成本，來提高運維效率。","attrs":{}}]},{"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":"比如剛纔提到5G能耗問題，我們能否通過人工智能的技術來去降低能耗呢？從過往的實踐經驗來看，上述問題的答案是肯定的。","attrs":{}}]},{"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":"接下來，我們通過三個例子來說明。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/89/89603ebe1e5a9fc3433928e1228b17f9.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":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"1. 基站節能","attrs":{}}]},{"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":"第一個例子是基站節能。基站的能耗是非常高的。在佈網初期，基站用戶較少，有時候基站常常是空開。針對這種情況，運營商的解決方案是對話務量做出一些預測。如果我們能精準預測話務量的話，那麼，在話務量小的時候，我們就可以把一定量的載波關掉，從而達到節能的目的。據統計，在預測話務量的過程中，通過LSTM神經網絡來做預測，可以實現節能10%以上。","attrs":{}}]},{"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","attrs":{}}],"text":"2. 核心網KPI異常檢測","attrs":{}}]},{"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":"第二個例子，是異常檢測。在運營商的核心網部署KPI異常檢測服務。原有的異常檢測服務，是使用固定閾值進行告警通知。而AI技術，則更加智能、及時、準確地識別異常。","attrs":{}}]},{"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","attrs":{}}],"text":"3. 故障識別及根因定位","attrs":{}}]},{"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":"通常網絡上一旦發生故障，就會觸發大量的告警，而系統同時又以高經緯維度進行運維派單。如果多個網員上報多個告警，那麼就會出現這種重複派單。也就是說發生了一個故障，多網員上報告警，最後可能導致在多個域（無線域和傳輸域等）都去派單。","attrs":{}}]},{"type":"heading","attrs":{"align":null,"level":2},"content":[{"type":"text","text":"（三）開發AI應用仍然面臨挑戰：開發門檻高、週期長","attrs":{}}]},{"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":"從上面三個例子我們可以看出，AI相對來說，還是非常靠譜的。但是既然AI如此靠譜，爲什麼沒有得到全面快速的應用呢？因爲AI的開發還面臨着不小的挑戰，簡單概括就是六個字：","attrs":{}},{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"門檻高，週期長","attrs":{}},{"type":"text","text":"。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/4b/4b06784ec3ecbd5c15aa4f427ea4979f.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":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"​上圖是Gartner的一份研究報告。它從四個維度分析了AI應用的主要障礙。其中最主要的3點：","attrs":{}}]},{"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":"人員技能","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"理解增益與用途","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"數據範圍與質量","attrs":{}}]}]}],"attrs":{}},{"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":"這就回到我們說的六個字：門檻高，週期長。","attrs":{}}]},{"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","attrs":{}}],"text":"1. 門檻高","attrs":{}}]},{"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":"此處說的“門檻高”，","attrs":{}},{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"第一點是指缺乏AI算法開發人員。","attrs":{}},{"type":"text","text":"一般的運維團隊不會配置專門的AI算法開發人員，這樣必然導致AI技能的缺失。","attrs":{}}]},{"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":"但這不是最關鍵的，因爲AI人員通過培訓、培養、招聘等手段，都可以解決。 ","attrs":{}}]},{"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":"最關鍵的，","attrs":{}},{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"也就是我們說的第二點，算法與業務結合難。","attrs":{}},{"type":"text","text":"如果要想把一個應用做好，最好的是從業務出發，根據業務的實際情況選擇合適的算法，這樣才能把應用做好。但在實際操作過程中，首先，我們需要有一個業務專家對運維要有深刻的理解；其次，還需要有一個精通AI的算法專家。在這之後，需要他們有充足的時間和意願坐下來深入的交流。在這裏，時間和意願都會成爲阻礙。","attrs":{}}]},{"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","attrs":{}}],"text":"第三點是數據。","attrs":{}},{"type":"text","text":"數據包含兩個問題：工程問題和標註問題。即，開發一個AI應用實際上是相當大的工程量，因爲首先需要接入海量的多模態的數據去完成模型的訓練和推理，最後還要去完成結果的展示，包括去對接一些現有的系統。因此除了前面需要的運維專家和算法專家，還需要很多工程開發人員。","attrs":{}}]},{"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","attrs":{}}],"text":"2. 週期長","attrs":{}}]},{"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":"開發門檻高，就決定了開發週期長，畢竟有這麼高的門檻，如果不能很好的解決的話，那麼週期必然會特別長。開發週期長會導致：","attrs":{}}]},{"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":"第一，","attrs":{}},{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"理解增益和用途","attrs":{}},{"type":"text","text":"。怎麼理解呢？也就是說，如果我們長時間拿不到結果，那麼企業決策人員就可能對AI能產生的效果會表示懷疑；","attrs":{}}]},{"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":"第二，","attrs":{}},{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"時間越長，大家對項目的期望就會越高。","attrs":{}},{"type":"text","text":"假設同樣是做一個東西取得了同樣的效果，比如說故障修復時長降低5%，兩年做出來的和一個月做出來的，得到的評價可能就完全不一樣。","attrs":{}}]},{"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":"針對AIOps落地過程中遇到的挑戰，華爲推出的AIOps服務！現在我們一起來看看AIOps服務具體是什麼，以及它是如何解決我們前面面臨的挑戰的。","attrs":{}}]},{"type":"heading","attrs":{"align":null,"level":1},"content":[{"type":"text","text":"華爲AIOps服務","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/ce/ce3457e038b2a995c3240fa42ef4d344.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":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"​上圖是AIOps服務的整體框架。AIOps從下到上分成了四層：","attrs":{}}]},{"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","attrs":{}}],"text":"第一層：數據的採集和治理。","attrs":{}},{"type":"text","text":"數據採集治理，聽上去容易，做起來難，爲什麼呢？因爲要面對的數據類型多，接口和數據類型也不統一。光去適配這些數據，都有可能累的焦頭爛額。相對來說，華爲AIOps服務首先支持通用的接口，然後對一些常見的設備都已經預置完成，最後能達到自動對接，數據自動治理的一個水平。","attrs":{}}]},{"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","attrs":{}}],"text":"第二層：AI原子能力。","attrs":{}},{"type":"text","text":"華爲AIOps共有二十多個原子能力，覆蓋檢測、預測、識別、診斷四大場景。原子能力不僅僅是AI算法的一個實現。每一個原子能力都經過實際局點數據的檢驗，針對具體的運營場景做過優化。同時，每一個原子能力也都融入了華爲以前的運維經驗，某些原子能力甚至能做到不訓練可以直接使用。 ","attrs":{}}]},{"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","attrs":{}}],"text":"第三層：編排能力。","attrs":{}},{"type":"text","text":"包括流程的編排和大屏的編排，還有RPA的編排。原子能力是AIOps智能運維的基礎組件，流程編排操作簡單靈活，只需從組件庫中拖拽數據及AI運維能力進行組合，即可完成命令場景端到端的圖形化編排，真正支撐合作伙伴拉低開發門檻，高效率的構建AI應用編排框架。","attrs":{}}]},{"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","attrs":{}}],"text":"第四層：行業AI app。","attrs":{}},{"type":"text","text":"針對最典型的場景開箱即用。通過豐富的2D和3D可視化組件，如提供了超過30個圖表控件，覆蓋折線、拓撲、列表、柱形等樣式，並提供多個地圖控件、交互控件及媒體控件搭建。運維效果大屏時只需從組件庫裏拖拽出各類控件，按需組合自由佈局、靈活配置應用的各種報表，輔助監控和分析，例如DIY微服務健康監控大廳，使其能夠可視化，展示接口平均成功率、接口平均時延、接口失敗率、接口調用次數等。同時提供KPI告警列表，爲運營人員提供故障預警參考依據，拖拽所需控件號，對控件的樣式，數據及交互進行個性化定製，使其滿足展示要求。後端數據還可使用app組合流程裏定義的各類中間數據。配置完成後即可一鍵預覽和發佈運維效果，大屏展示接口，平均成功率，接口平均時延，接口失敗率，接口調用次數等，快速實現DIY可視化大屏。","attrs":{}}]},{"type":"heading","attrs":{"align":null,"level":2},"content":[{"type":"text","text":"（一）RPA助力AIOps對接現有運維繫統","attrs":{}}]},{"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":"除了展示位，推理結果必須能夠幫助進行故障的恢復。現階段一般是對接現有的系統，比如工單系統（需要工單郵箱的人要去處理）、自動回覆和問題單。如果通過人工去對接，費時費力並且容易出錯。因此機器人流程自動化，也就是RPA服務，水到渠成。RPA服務可以完成數據的對接、搬運及工單的發放等等，減少人力投入，降低出錯成本。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/f6/f6be5dec2ddc60bc51ee5b8bec8fff9b.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":null,"origin":null}},{"type":"heading","attrs":{"align":null,"level":2},"content":[{"type":"text","text":"（二）10+開箱即用的App，支持快速部署","attrs":{}}]},{"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":"針對一些最典型的場景，華爲雲AIOps把編排能力都已經提前準備好，也即，","attrs":{}},{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"有十多種開箱即用的App","attrs":{}},{"type":"text","text":"，如園區網絡、DC網絡、IT應用、運營商網絡等等場景全覆蓋；","attrs":{}},{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"靈活部署","attrs":{}},{"type":"text","text":"，支持公有云、HCS部署、On Premise部署、及雲地協同等；","attrs":{}},{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"開放生態，","attrs":{}},{"type":"text","text":"支持合作伙伴開發行業App，並將AI應用發佈到AI市場，合作共贏，共建網絡AI生態。","attrs":{}}]},{"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":"下面我們以“KPI異常檢測”App來演示一下如何使用一個開箱即用的App。","attrs":{}}]},{"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":"第一步：導入網元列表；","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"第二步：配置性能、告警數據源；","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"第三步：數據源關聯到App；","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"第四步：啓動App；","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"第五步：查看大屏，分析故障。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/f3/f37fb53ea35f848ea1fa758c7e360efb.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":null,"origin":null}},{"type":"heading","attrs":{"align":null,"level":1},"content":[{"type":"text","text":"AIOps使能園區網絡智能運維","attrs":{}}]},{"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":"那麼AIOps是如何解決園區中實際運維的呢？","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"heading","attrs":{"align":null,"level":2},"content":[{"type":"text","text":"（一）園區網絡建維模式","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/11/11f564be05e1d494afde4425b6dc0e4e.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":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"​上圖爲園區網絡的兩種建維模式：","attrs":{}}]},{"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","attrs":{}}],"text":"2B和2C共用大網的OMC：","attrs":{}},{"type":"text","text":"當前的主流模式。企業去租用運營商的無線設備及其他的一些設備。這種模式的問題在於，終端由企業維護，網絡由運營商維護，那麼出現問題的時候很難分清責任；另外一個問題是，運營商側的運維能力和組織構築大網2C的O域，難以支撐企業內網高SLA，強化客戶訴求。","attrs":{}}]},{"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","attrs":{}}],"text":"2B和2C分開OMC（EMS）：","attrs":{}},{"type":"text","text":"企業採購5G CPE、無線、核心網等全部設備進行維護，具備端到端的視圖。從工信部發文、VDF、奧迪園區及企業SLA保障來看，","attrs":{}},{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"企業租用運營商頻譜或專用頻譜自建5G網絡會逐步成爲主流。","attrs":{}}]},{"type":"heading","attrs":{"align":null,"level":2},"content":[{"type":"text","text":"（二）業務場景和痛點分析：園區客戶需要簡單易用、多域融合的網絡運維","attrs":{}}]},{"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","attrs":{}}],"text":" 1. 典型網絡現狀","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/67/676d28c3285993a2df215aff2284d162.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":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"​上圖是一個園區比較常見的一個視頻檢測的業務。我們可以看到，即便是一個最常見的業務，也大概十來個網元都會參與到其中，從5G的無線到傳輸到邊緣計算，甚至是核心網，都會去參與其中。","attrs":{}}]},{"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","attrs":{}}],"text":"2. 園區應用","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/79/79f1cfbe4a41ee377a17b586b6ab31ae.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":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"​上圖列出了園區裏面常見的一些應用，包括邊緣的AI檢測、智能物流、室內定位等。所有的這些業務其實都和上一張圖類似，即任何一個簡單的業務都要涉及到多個域的參與。","attrs":{}}]},{"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":"那麼園區與運營商運維的差異是什麼呢？主要有以下三點：","attrs":{}}]},{"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","attrs":{}}],"text":"用戶：","attrs":{}},{"type":"text","text":"缺乏專業的通信知識，網絡運維能力弱；","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"網絡：","attrs":{}},{"type":"text","text":"組網相對簡單，但涉及多域、無線、傳接、數通、IT等；","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"SLA：","attrs":{}},{"type":"text","text":"生產系統網絡端到端SLA合同要求高，7X24小時，99.99%。","attrs":{}}]},{"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":"因此，客戶如果是園區運維的話，有如下痛點：","attrs":{}}]},{"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","attrs":{}}],"text":"技能：","attrs":{}},{"type":"text","text":"5G 2B引入使得網絡更加複雜，企業工程師缺乏相關技能，運維困難；","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"工具：","attrs":{}},{"type":"text","text":"缺乏有效的運維工具，複雜網絡問題定位需要跨域專家現場會診，成本高，耗時長。","attrs":{}}]},{"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":"總結來說","attrs":{}},{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"，園區網絡跨域設備需要實現數據融合，支撐端到端分析及呈現，最終實現企業ICT基礎設施的統一運維。而園區網絡涉及網絡設備多，邊界模糊，需要有統一的跨域定界定位能力，加速生產網絡問題定位。","attrs":{}}]},{"type":"heading","attrs":{"align":null,"level":2},"content":[{"type":"text","text":"（三）傳統人工、工具化運維不能滿足園區網絡新需求，急需智能化轉型","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/66/66f07a0377c339464d7f906a879c2a55.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":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"​根據上圖的數據，我們可以看到：","attrs":{}}]},{"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","attrs":{}}],"text":"被動式運維：","attrs":{}},{"type":"text","text":"75%的問題都是由用戶發現而非主動檢測，如果由用戶發現，那麼用戶很可能就會投訴；","attrs":{}}]},{"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","attrs":{}}],"text":"自動化程度低：","attrs":{}},{"type":"text","text":"企業成本中70%的運營成本屬於人力成本，成本激增；","attrs":{}}]},{"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","attrs":{}}],"text":"故障解決困難：","attrs":{}},{"type":"text","text":"90%故障的恢復時間是用來做問題定位的，真正的問題修復時間佔比非常小。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":" ","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"這樣看來，","attrs":{}},{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"無論是從效率還是效果這兩方面去考慮，都有一個訴求就是引入人工智能去解決問題，使能網絡運維的預測、分析、決策的自動化閉環","attrs":{}},{"type":"text","text":"。","attrs":{}}]},{"type":"heading","attrs":{"align":null,"level":2},"content":[{"type":"text","text":"（四）跨域故障定位算法流程","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/d5/d56664c514ab9b74da9e1ec489bbb2a1.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":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"​上圖是跨域故障定位的算法流程。整個流程如下：","attrs":{}}]},{"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","attrs":{}}],"text":"輸入：","attrs":{}}]},{"type":"bulletedlist","content":[{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"告警：設備上報的告警；","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"Topo：組網Topo結構；","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"故障傳播圖：告警間的影響關係。","attrs":{}}]}]}],"attrs":{}},{"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","attrs":{}}],"text":"流程介紹：","attrs":{}}]},{"type":"bulletedlist","content":[{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"降噪：過濾原始告警中的閃斷、震斷等數量多又無效告警；","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"聚合：對告警進行劃分，將Topo不相關的告警分開，可能相關（屬於同一故障）的告警聚合到一起，得到多個告警組；","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"識別定位：結合Topo、故障傳播圖，對每個告警組進行分析，識別出每個告警組中有幾個故障，每個故障的根因網元和根因告警；","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"診斷：對於每個故障告警診斷出故障的類型，例如：電源中斷。","attrs":{}}]}]}],"attrs":{}},{"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","attrs":{}}],"text":"輸出：","attrs":{}}]},{"type":"bulletedlist","content":[{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"故障的根因","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"故障設計的告警","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"故障類型","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"故障恢復建議","attrs":{}}]}]}],"attrs":{}},{"type":"heading","attrs":{"align":null,"level":2},"content":[{"type":"text","text":"（五）AIOps框架實現算法流程","attrs":{}}]},{"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":"以上講解了整個的算法流程，接下來，我們看看如果使用華爲AIOps框架去實現算法流程。","attrs":{}}]},{"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","attrs":{}}],"text":"1、快速配置數據源，編排流程","attrs":{}}]},{"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":"配置數據源：將無線、傳輸、核心網等多個域的告警接入，接入網絡拓撲數據；","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"流程編排：通用已有的原子能力，快速進行流程編排。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/e0/e0020afef040fab0fabe69104dfd1f9a.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":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"​經過上述過程，可以完成“事件通知”功能，並將結果保存到記錄集（即，數據庫），用於大屏展示。效果圖如下：","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/96/9648304bec083c3e6d8230d2e6aee09f.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":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"打開其中一條告警，可以看到如下信息：","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/bc/bc7b0928cdde2701d53815b2e965e9eb.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":null,"origin":null}},{"type":"heading","attrs":{"align":null,"level":1},"content":[{"type":"text","text":"AIOps部署建議","attrs":{}}]},{"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":"根據前述的實踐，我們可以總結以下內容：","attrs":{}}]},{"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","attrs":{}}],"text":"1、選定成熟場景，循序漸進部署AIOps","attrs":{}}]},{"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":"經過長期實踐，我們對AIOps部署失敗的主要原因做了如下總結：","attrs":{}}]},{"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","attrs":{}}],"text":"數據上不來：","attrs":{}},{"type":"text","text":"數據分散在各個獨立系統之上，缺乏綜合採集管理手段。數據缺失，數據質量低下是造成AIOps效果欠佳的主要原因；","attrs":{}}]},{"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","attrs":{}}],"text":"命令下不去：","attrs":{}},{"type":"text","text":"缺乏自動化運維工具，不能進行主動檢測，恢復操作；","attrs":{}}]},{"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","attrs":{}}],"text":"模型不智能：","attrs":{}},{"type":"text","text":"不能有效的積累日常運維中的標註信息，不能實現模型自學習。","attrs":{}}]},{"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":"因此，在部署失敗的基礎上，我們可以得出，如果要成功部署AIOps，我們需要：","attrs":{}}]},{"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":"從具備條件的","attrs":{}},{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"成熟場景","attrs":{}},{"type":"text","text":"出發，","attrs":{}},{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"循序漸進","attrs":{}},{"type":"text","text":"推進AIOps部署；","attrs":{}}]},{"type":"bulletedlist","content":[{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"數據上的來","attrs":{}},{"type":"text","text":"，全面收集各種運維數據，提高數據質量；","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"命令下得去","attrs":{}},{"type":"text","text":"，AIOps後端對接現在自動運維工具，增強診斷手段和自動恢復能力；","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"有效積累標註數據","attrs":{}},{"type":"text","text":"，讓AIOps模型能不斷收到反饋，具備自學習能力。","attrs":{}}]}]}],"attrs":{}},{"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","attrs":{}}],"text":"2、選擇成熟的AIOps服務","attrs":{}}]},{"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":"針對不同類型的企業，AIOps服務的選擇也是不盡相同，具體見下表：","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/ac/ac6a831e2e121d12ed105fe3147a3cfa.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":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/d5/d5559c6065a3a35a79e221e197b5332b.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":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"​華爲AlOps服務降低網絡AI應用開發門檻，加速網絡AI應用落地。沉澱了10+開箱即用的智能APP，覆蓋運營商網絡、園區網絡、數據中心網絡和IT應用等應用領域。預集成豐富的AI原子能力，覆蓋故障預測、檢測、診斷、識別等環節。支持用戶零編碼開發AI應用，提升運維效率。","attrs":{}}]},{"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":"感興趣就一起來體驗一下吧~ ","attrs":{}},{"type":"link","attrs":{"href":"https://www.hwtelcloud.com/products/aio","title":"","type":null},"content":[{"type":"text","text":"https://www.hwtelcloud.com/products/aio","attrs":{}}]}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"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://bbs.huaweicloud.com/blogs?utm_source=infoq&utm_medium=bbs-ex&utm_campaign=other&utm_content=content","title":"","type":null},"content":[{"type":"text","text":"點擊關注，第一時間瞭解華爲雲新鮮技術~","attrs":{}}]}]}]}