自动驾驶发展_自动驾驶网络及其发展
自動(dòng)駕駛發(fā)展
介紹 (Introduction)
Talking about inspiration in the networking industry, nothing more than Autonomous Driving Network (ADN). You may hear about this and wondering what this is about, and does it have anything to do with autonomous driving vehicles? Your guess is right; the ADN concept is derived from or inspired by the rapid development of the autonomous driving car in recent years.
談到網(wǎng)絡(luò)行業(yè)的靈感，無非就是自動(dòng)駕駛網(wǎng)絡(luò)(ADN)。 您可能聽說過這，并想知道這是什么，它與自動(dòng)駕駛汽車有什么關(guān)系嗎？ 您的猜測(cè)是正確的； ADN概念源于自動(dòng)駕駛汽車近年來的快速發(fā)展或受其啟發(fā)。
Driverless Car of the Future, the advertisement for “America’s Electric Light and Power Companies,” Saturday Evening Post, the 1950s. Credit: The Everett Collection. (Mark W., 2014)未來無人駕駛汽車，“美國(guó)的電燈和電力公司”的廣告，1950年代，星期六晚上郵報(bào)。 信用：Everett收藏。 (馬克·W，2014年)The vision of autonomous driving has been around for more than 70 years. But engineers continuously make attempts to achieve the idea without too much success. The concept stayed as a fiction for a long time. In 2004, the US Defense Advanced Research Projects Administration (DARPA) organized the Grand Challenge for autonomous vehicles for teams to compete for the grand prize of $1 million. I remembered watching TV and saw those competing vehicles, behaved like driven by drunk man, had a really tough time to drive by itself. I thought that autonomous driving vision would still have a long way to go. To my surprise, the next year, 2005, Stanford University’s vehicles autonomously drove 131 miles in California’s Mojave desert without a scratch and took the $1 million Grand Challenge prize. How was that possible? Later I learned that the secret ingredient to make this possible was using the latest ML (Machine Learning) enabled AI (Artificial Intelligent ) technology.
自動(dòng)駕駛的愿景已經(jīng)存在了70多年了。 但是工程師們不斷地嘗試實(shí)現(xiàn)這一想法，但并沒有取得太大的成功。 這個(gè)概念長(zhǎng)期以來一直是虛構(gòu)的。 2004年，美國(guó)國(guó)防高級(jí)研究計(jì)劃局(DARPA)組織了無人駕駛汽車大挑戰(zhàn)，各車隊(duì)爭(zhēng)奪100萬(wàn)美元的大獎(jiǎng)。 我記得看電視，看到那些競(jìng)爭(zhēng)的車輛，表現(xiàn)得像醉漢一樣，很難開車。 我認(rèn)為自動(dòng)駕駛的視野還很長(zhǎng)。 令我驚訝的是，第二年，2005年，斯坦福大學(xué)的汽車在加州莫哈韋沙漠無人駕駛自動(dòng)行駛131英里，并獲得了100萬(wàn)美元的“挑戰(zhàn)大獎(jiǎng)”。 那怎么可能？ 后來我得知，實(shí)現(xiàn)這一目標(biāo)的秘密因素是使用最新的ML(機(jī)器學(xué)習(xí))支持的AI(人工智能)技術(shù)。
Since then, AI technologies advanced rapidly and been implemented in all verticals. Around the 2016 time frame, the concept of Autonomous Driving Network started to emerge by combining AI and network to achieve network operational autonomy. The automation concept is nothing new in the networking industry; network operations are continually being automated here and there. But this time, ADN is beyond automating mundane tasks; it reaches a whole new level. With the help of AI technologies and other critical ingredients advancement like SDN (Software Defined Network), autonomous networking has a great chance from a vision to future reality.
從那時(shí)起，人工智能技術(shù)飛速發(fā)展，并在各個(gè)領(lǐng)域得到了實(shí)現(xiàn)。 在2016年左右的時(shí)間里，自動(dòng)駕駛網(wǎng)絡(luò)的概念開始出現(xiàn)，它通過將AI和網(wǎng)絡(luò)相結(jié)合來實(shí)現(xiàn)網(wǎng)絡(luò)運(yùn)營(yíng)自主性。 自動(dòng)化概念在網(wǎng)絡(luò)行業(yè)中并不是什么新鮮事物。 網(wǎng)絡(luò)操作在這里和那里不斷地自動(dòng)化。 但是這次，ADN超出了自動(dòng)執(zhí)行日常任務(wù)的范圍； 它達(dá)到了一個(gè)全新的水平。 借助AI技術(shù)和SDN(軟件定義網(wǎng)絡(luò))等其他關(guān)鍵要素的進(jìn)步，從愿景到未來現(xiàn)實(shí)，自主網(wǎng)絡(luò)都是一個(gè)巨大的機(jī)會(huì)。
In this article, we will examine some critical components of the ADN, current landscape, and factors that are important for ADN to be a success.
在本文中，我們將研究ADN的一些關(guān)鍵組成部分，當(dāng)前情況以及對(duì)ADN成功至關(guān)重要的因素。
愿景 (The Vision)
At the current stage, there are different terminologies to describe ADN vision by various organizations.
在當(dāng)前階段，有各種術(shù)語(yǔ)來描述各個(gè)組織的ADN愿景。
Even though slightly different terminologies, the industry is moving towards some common terms and consensus called autonomous networks, e.g. TMF, ETSI, ITU-T, GSMA. The core vision includes business and network aspects. The autonomous network delivers the “hyper-loop” from business requirements all the way to network and device layers.
盡管術(shù)語(yǔ)略有不同，但業(yè)界仍在朝著一些稱為自治網(wǎng)絡(luò)的通用術(shù)語(yǔ)和共識(shí)邁進(jìn)，例如TMF，ETSI，ITU-T，GSMA。 核心愿景包括業(yè)務(wù)和網(wǎng)絡(luò)方面。 自治網(wǎng)絡(luò)從業(yè)務(wù)需求一直到網(wǎng)絡(luò)和設(shè)備層都提供“超環(huán)”。
On the network layer, it contains the below critical aspects:
在網(wǎng)絡(luò)層，它包含以下關(guān)鍵方面：
Intent-Driven: Understand the operator’s business intent and automatically translate it into necessary network operations. The operation can be a one-time operation like disconnect a connection service or continuous operations like maintaining a specified SLA (Service Level Agreement) at the all-time.
目的驅(qū)動(dòng)：了解運(yùn)營(yíng)商的業(yè)務(wù)意圖并將其自動(dòng)轉(zhuǎn)換為必要的網(wǎng)絡(luò)操作。 該操作可以是一次性操作，例如斷開連接服務(wù)，也可以是連續(xù)操作，例如始終保持指定的SLA(服務(wù)水平協(xié)議)。
Self-Discover: Automatically discover hardware/software changes in the network and populate the changes to the necessary subsystems to maintain always-sync state.
自我發(fā)現(xiàn)：自動(dòng)發(fā)現(xiàn)網(wǎng)絡(luò)中的硬件/軟件更改，并將更改填充到必要的子系統(tǒng)中，以保持始終同步狀態(tài)。
Self-Config/Self-Organize: Whenever network changes happen, automatically configure corresponding hardware/software parameters such that the network is at the pre-defined target states.
自我配置/自我組織：每當(dāng)網(wǎng)絡(luò)發(fā)生變化時(shí)，都將自動(dòng)配置相應(yīng)的硬件/軟件參數(shù)，以使網(wǎng)絡(luò)處于預(yù)定義的目標(biāo)狀態(tài)。
Self-Monitor: Constantly monitor networks/services operation states and health conditions automatically.
自我監(jiān)控：不斷自動(dòng)監(jiān)控網(wǎng)絡(luò)/服務(wù)的運(yùn)行狀態(tài)和健康狀況。
Auto-Detect: Detect network faults, abnormalities, and intrusions automatically.
自動(dòng)檢測(cè)：自動(dòng)檢測(cè)網(wǎng)絡(luò)故障，異常和入侵。
Self-Diagnose: Automatically conduct an inference process to figure out the root causes of issues.
自我診斷：自動(dòng)進(jìn)行推理過程以找出問題的根本原因。
Self-Healing: Automatically take necessary actions to address issues and bring the networks/services back to the desired state.
自我修復(fù)：自動(dòng)采取必要的措施來解決問題并將網(wǎng)絡(luò)/服務(wù)恢復(fù)到所需狀態(tài)。
Self-Report: Automatically communicate with its environment and exchange necessary information.
自我報(bào)告：自動(dòng)與其環(huán)境進(jìn)行通信并交換必要的信息。
Automated common operational scenarios: Automatically perform operations like network planning, customer and service onboarding, network change management.
自動(dòng)化的常見操作場(chǎng)景：自動(dòng)執(zhí)行網(wǎng)絡(luò)規(guī)劃，客戶和服務(wù)入門，網(wǎng)絡(luò)變更管理等操作。
On top of those, these capabilities need to be across multiple services, multiple domains, and the entire lifecycle(TMF, 2019).
最重要的是，這些功能需要跨多個(gè)服務(wù)，多個(gè)域以及整個(gè)生命周期(TMF，2019)。
No doubt, this is the most ambitious goal that the networking industry has ever aimed at. It has been described as the “end-state” and“ultimate goal” of networking evolution. This is not just a vision on PPT, the networking industry already on the move toward the goal.
毫無疑問，這是網(wǎng)絡(luò)行業(yè)有史以來最雄心勃勃的目標(biāo)。 它被描述為網(wǎng)絡(luò)發(fā)展的“最終狀態(tài)”和“最終目標(biāo)”。 這不僅僅是關(guān)于PPT的愿景，網(wǎng)絡(luò)行業(yè)已經(jīng)朝著目標(biāo)邁進(jìn)。
David Wang, Huawei’s Executive Director of the Board and President of Products & Solutions, said in his 2018 Ultra-Broadband Forum(UBBF) keynote speech. (David W. 2018):
華為執(zhí)行董事兼產(chǎn)品與解決方案總裁王大衛(wèi)在2018年超寬帶論壇(UBBF)主題演講中表示。 (David W.2018)：
“In a fully connected and intelligent era, autonomous driving is becoming a reality. Industries like automotive, aerospace, and manufacturing are modernizing and renewing themselves by introducing autonomous technologies. However, the telecom sector is facing a major structural problem: Networks are growing year by year, but OPEX is growing faster than revenue. What’s more, it takes 100 times more effort for telecom operators to maintain their networks than OTT players. Therefore, it’s imperative that telecom operators build autonomous driving networks.”
在完全連接和智能化的時(shí)代，自動(dòng)駕駛已成為現(xiàn)實(shí)。 汽車，航空航天和制造業(yè)等行業(yè)正在通過引入自主技術(shù)來實(shí)現(xiàn)自我更新和更新。 但是，電信行業(yè)面臨著一個(gè)主要的結(jié)構(gòu)性問題：網(wǎng)絡(luò)每年都在增長(zhǎng)，但是運(yùn)營(yíng)支出的增長(zhǎng)快于收入。 而且，與OTT運(yùn)營(yíng)商相比，電信運(yùn)營(yíng)商維護(hù)其網(wǎng)絡(luò)所花費(fèi)的精力要多100倍。 因此，電信運(yùn)營(yíng)商必須建立自動(dòng)駕駛網(wǎng)絡(luò)。”
Juniper CEO Rami Rahim said in his keynote at the company’s virtual AI event: (CRN, 2020)
瞻博網(wǎng)絡(luò)首席執(zhí)行官拉米·拉希姆(Rami Rahim)在公司虛擬AI活動(dòng)的主題演講中說： (CRN，2020)
“The goal now is a self-driving network. The call to action is to embrace the change. We can all benefit from putting more time into higher-layer activities, like keeping distributors out of the business. The future, I truly believe, is about getting the network out of the way. It is time for the infrastructure to take a back seat to the self-driving network.”
“現(xiàn)在的目標(biāo)是建立自動(dòng)駕駛網(wǎng)絡(luò)。 呼吁采取行動(dòng)就是擁抱變化。 將更多的時(shí)間投入到更高層次的活動(dòng)中，例如使分銷商脫離業(yè)務(wù)，我們都可以從中受益。 我真正相信，未來將使網(wǎng)絡(luò)暢通無阻。 現(xiàn)在該讓基礎(chǔ)架構(gòu)在自動(dòng)駕駛網(wǎng)絡(luò)中退后一步了。”
這個(gè)愿景可以實(shí)現(xiàn)嗎？ (Is This Vision Achievable?)
If you asked me this question 15 years ago, my answer would be “no chance” as I could not imagine an autonomous driving vehicle was possible then. But now, the vision is not far-fetch anymore not only because of ML/AI technology rapid advancement but other key building blocks are made significant progress, just name a few key building blocks:
如果您在15年前問我這個(gè)問題，我的回答將是“沒有機(jī)會(huì)”，因?yàn)槲耶?dāng)時(shí)無法想象有可能駕駛自動(dòng)駕駛汽車。 但是現(xiàn)在，這個(gè)愿景不再遙不可及，不僅因?yàn)镸L / AI技術(shù)的飛速發(fā)展，而且其他關(guān)鍵構(gòu)建塊也取得了重大進(jìn)展，僅舉幾個(gè)關(guān)鍵構(gòu)建塊：
- software-defined networking (SDN) control 軟件定義網(wǎng)絡(luò)(SDN)控制
- industry-standard models and open APIs 行業(yè)標(biāo)準(zhǔn)模型和開放API
- Real-time analytics/telemetry 實(shí)時(shí)分析/遙測(cè)
- big data processing 大數(shù)據(jù)處理
- cross-domain orchestration 跨域編排
- programmable infrastructure 可編程基礎(chǔ)架構(gòu)
- cloud-native virtualized network functions (VNF) 云原生虛擬化網(wǎng)絡(luò)功能(VNF)
- DevOps agile development process DevOps敏捷開發(fā)流程
- everything-as-service design paradigm 一切即服務(wù)的設(shè)計(jì)范例
- intelligent process automation 智能過程自動(dòng)化
- edge computing 邊緣計(jì)算
- cloud infrastructure 云基礎(chǔ)設(shè)施
- programing paradigm suitable for building an autonomous system . i.e., teleo-reactive programs, which is a set of reactive rules that continuously sense the environment and trigger actions whose continuous execution eventually leads the system to satisfy a goal. (Nils Nilsson, 1996) 適合建立自治系統(tǒng)的程序設(shè)計(jì)范例。 即遠(yuǎn)程React程序，它是一組React性規(guī)則，可連續(xù)感知環(huán)境并觸發(fā)動(dòng)作，這些動(dòng)作的連續(xù)執(zhí)行最終使系統(tǒng)達(dá)到目標(biāo)。 (Nils Nilsson，1996年)
- open-source solutions 開源解決方案
巨大的挑戰(zhàn) (Huge Challenges)
We have reasons to be optimistic about ADN while fully realize the considerable challenges in this ADN journey.
我們有理由對(duì)ADN感到樂觀，同時(shí)充分意識(shí)到ADN旅程中的巨大挑戰(zhàn)。
As we know, a typical autonomous system composes of 3 essential components:
眾所周知，一個(gè)典型的自治系統(tǒng)包含3個(gè)基本組成部分：
An Agent: A reactive system (controller) interacting with components of its environment so that specific goals are met.
代理：React性系統(tǒng)(控制器)與其環(huán)境組件進(jìn)行交互，從而實(shí)現(xiàn)特定目標(biāo)。
An Object: A physical or virtual component whose behavior can be controlled by system agents.
對(duì)象： A 可以由系統(tǒng)代理控制其行為的物理或虛擬組件。
The Environment: Consists of the elements of the physical and virtual infrastructure of the system that is used for the coordination between components (agents and objects).
環(huán)境：由系統(tǒng)的物理和虛擬基礎(chǔ)結(jié)構(gòu)的元素組成，用于組件(代理和對(duì)象)之間的協(xié)調(diào)。
The first complexity of autonomous networking is a large number of objects (such as network elements, boards, links, and optical/electrical/wireless links). Each object has lots of tunable parameters and a lack of consistency of the object models. The other complexity is complex network environment: numerous networking protocols, operation models, layers of networking stacks, observability of the objects, communication infrastructures, and necessary interaction with the external environment like weather, sports events, and black swan events like current COVID19 pandemic. All these make building such an intelligent agent a very challenging task.
自主網(wǎng)絡(luò)的第一個(gè)復(fù)雜性是大量的對(duì)象(例如網(wǎng)絡(luò)元素，板，鏈路和光/電/無線鏈路)。 每個(gè)對(duì)象都有很多可調(diào)參數(shù)，并且對(duì)象模型缺乏一致性。 另一個(gè)復(fù)雜性是復(fù)雜的網(wǎng)絡(luò)環(huán)境：眾多的網(wǎng)絡(luò)協(xié)議，操作模型，網(wǎng)絡(luò)堆棧層，對(duì)象的可觀察性，通信基礎(chǔ)結(jié)構(gòu)以及與外部環(huán)境(如天氣，體育賽事和黑天鵝事件，如當(dāng)前的COVID19大流行)的必要交互。 所有這些使構(gòu)建這樣的智能代理成為一項(xiàng)非常艱巨的任務(wù)。
Some other significant challenges come from the fact that telecom networks, like other infrastructure systems, evolve over a long time and not have many opportunities to be built “from scratch.” Instead, the new or changed elements are always needed to fit into the previously made infrastructure. Why is the realization of autonomous driving vehicles so much painful comparing with that of the autonomous driving airplanes? Because it has to fit into the existing complex and never perfect road infrastructure. Same for ADN, existing segregated, ultra-complex, and less-perfect network infrastructure presents significant challenges for the industry.
其他一些重大挑戰(zhàn)來自這樣一個(gè)事實(shí)，即電信網(wǎng)絡(luò)與其他基礎(chǔ)架構(gòu)系統(tǒng)一樣，會(huì)長(zhǎng)期演進(jìn)，并且沒有很多“從頭開始”構(gòu)建的機(jī)會(huì)。 取而代之的是，總是需要新的或更改的元素以適應(yīng)先前制作的基礎(chǔ)結(jié)構(gòu)。 為什么與無人駕駛飛機(jī)相比，實(shí)現(xiàn)無人駕駛汽車如此痛苦？ 因?yàn)樗仨氝m合現(xiàn)有的復(fù)雜且永遠(yuǎn)都不完美的道路基礎(chǔ)設(shè)施。 與ADN相同，現(xiàn)有的隔離，超復(fù)雜且性能較差的網(wǎng)絡(luò)基礎(chǔ)結(jié)構(gòu)為行業(yè)帶來了巨大挑戰(zhàn)。
What is the strategy to get to ADN vision? As usual, the answer: divide, evolve, and conquer.
實(shí)現(xiàn)ADN愿景的策略是什么？ 和往常一樣，答案是：分裂，發(fā)展和征服。
自治網(wǎng)絡(luò)級(jí)別定義 (Autonomous Network Levels Definition)
As the autonomous driving vehicle, networking companies and industry groups also defined ADN evolution milestones.
作為自動(dòng)駕駛汽車，網(wǎng)絡(luò)公司和行業(yè)組織也定義了ADN演進(jìn)的里程碑。
TMF Autonomous networks levels (TMF, 2019)TMF自治網(wǎng)絡(luò)級(jí)別(TMF，2019)Level 0 — manual management: The system delivers assisted monitoring capabilities, which means executing dynamic tasks manually.
級(jí)別0-手動(dòng)管理：系統(tǒng)提供輔助的監(jiān)視功能，這意味著手動(dòng)執(zhí)行動(dòng)態(tài)任務(wù)。
Level 1 — assisted management: The system executes a certain repetitive sub-task based on pre-configured to increase execution efficiency.
級(jí)別1-輔助管理：系統(tǒng)根據(jù)預(yù)先配置的內(nèi)容執(zhí)行某些重復(fù)的子任務(wù)，以提高執(zhí)行效率。
Level 2 — partial autonomous network: The system enables closed-loop O&M for specific units based on the AI model under certain external environments.
級(jí)別2-部分自治網(wǎng)絡(luò)：系統(tǒng)在某些外部環(huán)境下基于AI模型為特定單元啟用閉環(huán)運(yùn)維。
Level 3 — conditional autonomous network: Building on L2 capabilities, the system with awareness can sense real-time environmental changes. In certain network domains, optimize and adjust itself to the external environment to enable intent-based closed-loop management.
級(jí)別3-有條件的自治網(wǎng)絡(luò)：具有L2功能的系統(tǒng)可以感知實(shí)時(shí)的環(huán)境變化。 在某些網(wǎng)絡(luò)域中，優(yōu)化自身并適應(yīng)外部環(huán)境以啟用基于意圖的閉環(huán)管理。
Level 4 — high autonomous network: The system, building on L3 capabilities, enables, in a more complicated cross-domain environment, analyze and make the decision based on predictive or active closed-loop management of service and customer experience-driven networks.
級(jí)別4-高度自治的網(wǎng)絡(luò)：該系統(tǒng)基于L3功能，可以在更復(fù)雜的跨域環(huán)境中基于對(duì)服務(wù)和客戶體驗(yàn)驅(qū)動(dòng)的網(wǎng)絡(luò)的預(yù)測(cè)性或主動(dòng)閉環(huán)管理來分析和制定決策。
Level 5 — full autonomous network: This level is the ultimate goal for telecom network evolution. The system possesses closed-loop automation capabilities across multiple services, multiple domains, and the entire lifecycle, achieving autonomous networks.
級(jí)別5-完全自治的網(wǎng)絡(luò)：此級(jí)別是電信網(wǎng)絡(luò)演進(jìn)的最終目標(biāo)。 該系統(tǒng)具有跨多個(gè)服務(wù)，多個(gè)域以及整個(gè)生命周期的閉環(huán)自動(dòng)化功能，從而實(shí)現(xiàn)了自治網(wǎng)絡(luò)。
Cisco Digital Network Readiness Model (Cisco, 2020)思科數(shù)字網(wǎng)絡(luò)就緒模型(Cisco，2020) Juniper Self Driving Network Levels (K. Kompella, 2019)瞻博網(wǎng)絡(luò)自駕網(wǎng)絡(luò)水平(K.Kompella，2019)It is apparent that no standardized industry definition yet on the description of the levels. Different organizations have different focuses. But the main threads are similar:
顯然，關(guān)于級(jí)別的描述還沒有標(biāo)準(zhǔn)化的行業(yè)定義。 不同的組織有不同的重點(diǎn)。 但是主線程是相似的：
技術(shù)架構(gòu) (Technology Architecture)
A typical autonomous system has the following five complementary essential functions.
典型的自治系統(tǒng)具有以下五個(gè)互補(bǔ)的基本功能。
Perception, e.g., interpretation of stimuli, removing ambiguity from complex input data and determining relevant information;
感知，例如刺激的解釋，消除復(fù)雜輸入數(shù)據(jù)中的歧義并確定相關(guān)信息；
Reflection, e.g., building/updating a faithful environment run-time model from which strategies meeting the goals can be computed;
反思，例如，建立/更新一個(gè)忠實(shí)的環(huán)境運(yùn)行時(shí)模型，從中可以計(jì)算出達(dá)到目標(biāo)的策略；
Goal management, e.g., choosing among possible goals the most appropriate for a given configuration of the environment model;
目標(biāo)管理，例如，在可能的目標(biāo)中選擇最適合給定環(huán)境模型配置的目標(biāo)；
Planning to achieve a particular goal;
計(jì)劃實(shí)現(xiàn)特定目標(biāo)；
Self-awareness/adaptation, e.g., the ability to create.
自我意識(shí)/適應(yīng)能力，例如創(chuàng)造能力。
The below diagram characterizes how these five elements interact with each other to achieve autonomy.
下圖描述了這五個(gè)元素如何相互影響以實(shí)現(xiàn)自治。
The Concept of Autonomy — Architecture Characterizations (Joseph S, 2019)自治的概念-建筑特征(Joseph S，2019)While this architecture very well applies to ADN conceptually, ADN has its unique complexities and need special considerations:
盡管此體系結(jié)構(gòu)在概念上非常適用于ADN，但ADN具有其獨(dú)特的復(fù)雜性，并且需要特殊考慮：
Many Agents: Compare with a well-bounded, single-purpose system, i.e., robot/car/spaceship, networking more like a society. No single intelligent agent controls every aspect of the network but rather a collection of hierarchical smart agents that responsible for their subsystems, work in tandem.
許多特工 ：與功能強(qiáng)大的單一用途系統(tǒng)(例如，機(jī)器人/汽車/宇宙飛船)相比，網(wǎng)絡(luò)更像一個(gè)社會(huì)。 沒有單個(gè)智能代理控制網(wǎng)絡(luò)的每個(gè)方面，而是負(fù)責(zé)其子系統(tǒng)的一系列分層智能代理協(xié)同工作。
Intent-Driven: With the evolution to higher and higher levels in ADN, the human operators involve less and less in the actual network operation control loop. Instead, focus more and more on defining the business goal and processes “outside” or “on” the control loop. In order effectively to convey the business purpose to ADN, intent-driven interaction is one of the critical architectural considerations to separate the actual implementation from the request.
目的驅(qū)動(dòng)：隨著ADN的不斷發(fā)展，人工操作員越來越少地參與實(shí)際的網(wǎng)絡(luò)操作控制回路。 取而代之的是，越來越多地專注于定義業(yè)務(wù)目標(biāo)并在控制循環(huán)的“外部”或“外部”進(jìn)行處理。 為了有效地將業(yè)務(wù)目的傳達(dá)給ADN，意圖驅(qū)動(dòng)的交互是將實(shí)際實(shí)現(xiàn)與請(qǐng)求分開的關(guān)鍵體系結(jié)構(gòu)考慮之一。
Centralized ML/AI Capabilities: In ADN, humans need a platform to generate/train/apply/optimize/share AI models. A centralized platform can make these work more efficient.
集中的ML / AI功能：在ADN中，人類需要一個(gè)平臺(tái)來生成/訓(xùn)練/應(yīng)用/優(yōu)化/共享AI模型。 集中式平臺(tái)可以使這些工作更有效率。
Centralized Data Lake: Telecom network creates large amounts of data from various data sources. It is essential to have a centralized big data platform to collect, store, analyze, clean, filter data. Each subsystem can subscribe to the data it needs and make decisions by combining it with its local data.
集中式數(shù)據(jù)湖：電信網(wǎng)絡(luò)從各種數(shù)據(jù)源創(chuàng)建大量數(shù)據(jù)。 集中管理至關(guān)重要 大數(shù)據(jù)平臺(tái)收集，存儲(chǔ)，分析，清理，過濾數(shù)據(jù)。 每個(gè)子系統(tǒng)都可以訂閱其所需的數(shù)據(jù)，并通過將其與其本地?cái)?shù)據(jù)結(jié)合起來進(jìn)行決策。
Centralized Process Definition: Even though ML/AI will have the capability to decide the proper actions in particular scenarios, human operators still define the majority of the action sequence because not all the decision-making considerations are available for the machines to make a decision. A centralized process definition platform is highly desirable for operators to create/optimize/deploy workflows effectively.
集中的流程定義：即使ML / AI能夠決定特定場(chǎng)景下的適當(dāng)動(dòng)作，但是操作員仍然定義了大多數(shù)動(dòng)作順序，因?yàn)椴⒎撬袥Q策因素都可用于機(jī)器做出決定。 對(duì)于操作員而言，非常需要集中式的流程定義平臺(tái)來有效地創(chuàng)建/優(yōu)化/部署工作流。
ADN基礎(chǔ)架構(gòu) (ADN Infrastructure)
The networking industry and research communities have accepted the idea that the best approach to achieve ADN is through a set of single-domain autonomy plus cross-domain orchestration. Each single domain autonomy forms an AS(Autonomous System). Juniper calls the AS as “network bots” (K. Kompella,2018); others called it a “mini closed-loops.” This idea should have no surprise to everybody because it is very much in line with modern everything-as-a-service notations and famous cloud-native, microservice architecture. Powered by data, the AS possess more intelligence than regular microservices; it contains closed-loop control capabilities, understands the intent, performs the required operations, maintenance own health, and continuously monitor and adjust to ensure the target state is maintained.
網(wǎng)絡(luò)行業(yè)和研究界已經(jīng)接受了這樣的想法，即實(shí)現(xiàn)ADN的最佳方法是通過一組單域自治和跨域編排。 每個(gè)單個(gè)域自治都形成一個(gè)AS(自治系統(tǒng))。 Juniper將AS稱為“網(wǎng)絡(luò)機(jī)器人”(K. Kompella，2018年)； 其他人則稱其為“迷你閉環(huán)”。 這個(gè)想法對(duì)所有人都不應(yīng)該感到驚訝，因?yàn)樗浅７犀F(xiàn)代的“一切即服務(wù)”概念和著名的云原生微服務(wù)架構(gòu)。 由數(shù)據(jù)驅(qū)動(dòng)，AS具有比常規(guī)微服務(wù)更多的智能； 它包含閉環(huán)控制功能，了解意圖，執(zhí)行所需的操作，維護(hù)自己的健康狀況，并持續(xù)監(jiān)視和調(diào)整以確保維持目標(biāo)狀態(tài)。
The ASs run on an “ADN Infracture.” A good infrastructure lays a good foundation for ASs to thrive.
AS在“ ADN違規(guī)”上運(yùn)行。 良好的基礎(chǔ)架構(gòu)為AS的發(fā)展奠定了良好的基礎(chǔ)。
In the “ADN Solution White Paper, 2020”, Huawei outlines its ADN target architecture and product strategies.
在《 ADN解決方案白皮書，2020年 》中，華為概述了其ADN目標(biāo)架構(gòu)和產(chǎn)品策略。
Huawei ADN targeted architecture (Huawei ADN Solution White Paper, 2020)華為ADN目標(biāo)架構(gòu)(《華為ADN解決方案白皮書》，2020年)Huawei’s ADN target architecture covers all aspects of telecom networks: wireless, access, transport, optical, campus, and data centers. All three layers: NE, Network, and Cloud, embed with AI capability, which enables building ASs of different scales.
華為的ADN目標(biāo)架構(gòu)涵蓋了電信網(wǎng)絡(luò)的所有方面：無線，接入，傳輸，光纖，園區(qū)和數(shù)據(jù)中心。 三層：NE，網(wǎng)絡(luò)和云，均嵌入了AI功能，可構(gòu)建不同規(guī)模的AS。
On product strategies, Huawei’s iMaster NAIE (Network AI Engine) acts as the AI platform for ADN. AUTIN (Automation Intelligent) business process definition platform to help carriers to build visualized, automated, and intelligent capabilities for operations. Meanwhile, MAE and NCE are mobile and fix network management, control, and analytics platforms.
在產(chǎn)品策略上，華為的iMaster NAIE(網(wǎng)絡(luò)AI引擎)充當(dāng)ADN的AI平臺(tái)。 AUTIN(自動(dòng)化智能)業(yè)務(wù)流程定義平臺(tái)可幫助運(yùn)營(yíng)商構(gòu)建可視化，自動(dòng)化和智能化的運(yùn)營(yíng)能力。 同時(shí)，MAE和NCE是移動(dòng)的，可修復(fù)網(wǎng)絡(luò)管理，控制和分析平臺(tái)。
Huawei ADN System Panorama (Huawei ADN Solution White Paper, 2020)華為ADN系統(tǒng)全景圖(華為ADN解決方案白皮書，2020)This ADN infrastructure empowers AI anywhere with a clear demarcation of responsibilities. AI platform is especially important because of the skill gap among the architects who provide the design and requirements, the data scientists who built the models, and the software developers who develop and deploy end services.
這種ADN基礎(chǔ)結(jié)構(gòu)可通過明確劃分職責(zé)的方式在任何地方為AI提供支持。 由于提供設(shè)計(jì)和需求的架構(gòu)師，構(gòu)建模型的數(shù)據(jù)科學(xué)家以及開發(fā)和部署最終服務(wù)的軟件開發(fā)人員之間的技能差距，AI平臺(tái)尤其重要。
In the Open Source community, Linux Foundation’s ONAP (Open Network Automation Platform) also targeted as the infrastructure to realize ADN. The gap of missing AI platforms has been addressed by several new initiatives from the Linux Foundation, one being Project Acumos, supported by AT&T.
在開源社區(qū)中，Linux Foundation的ONAP(開放網(wǎng)絡(luò)自動(dòng)化平臺(tái))也以實(shí)現(xiàn)ADN的基礎(chǔ)結(jié)構(gòu)為目標(biāo)。 Linux基金會(huì)的多項(xiàng)新計(jì)劃已經(jīng)解決了AI平臺(tái)缺失的問題，其中一項(xiàng)是由AT＆T支持的Project Acumos 。
自治系統(tǒng)(AS) (Autonomous System (AS))
The many ASs work together in a defined ADN infrastructure to deliver overall ADN experience. How to identify and partition the responsibility of AS is a question that needs to consider. The AS can be packaged by technology, by region, by a legal entity, by functions, for re-use, for security, for simplified abstractions. Each company has its way of defining AS based on its own belief and available technologies.
許多AS在定義的ADN基礎(chǔ)結(jié)構(gòu)中一起工作，以提供整體ADN體驗(yàn)。 如何確定和劃分AS的職責(zé)是一個(gè)需要考慮的問題。 可以按技術(shù)，地區(qū)，法人實(shí)體，功能，重新使用，安全性，簡(jiǎn)化抽象的形式對(duì)AS進(jìn)行打包。 每個(gè)公司都有基于自己的信念和可用技術(shù)定義AS的方法。
If we want to achieve “open networking” for the benefits of sharing, simplifying, flexibility and cost-saving, standard organizations and open source communities can play significant roles in defining common ways to partition AS.
如果我們要實(shí)現(xiàn)“開放網(wǎng)絡(luò)”以共享，簡(jiǎn)化，靈活性和節(jié)省成本的優(yōu)勢(shì)，那么標(biāo)準(zhǔn)組織和開放源代碼社區(qū)可以在定義劃分AS的通用方法方面發(fā)揮重要作用。
The standard organizations are already making a move, e.g., ETSI published ZSM Requirements and Reference Architecture documents in 2019. ETSI defined a high-level architecture of ZSM. This ZSM framework would become the ADN infrastructure.
標(biāo)準(zhǔn)組織已經(jīng)在采取行動(dòng)，例如ETSI在2019年發(fā)布了ZSM需求和參考體系結(jié)構(gòu)文檔。ETSI定義了ZSM的高級(jí)體系結(jié)構(gòu)。 該ZSM框架將成為ADN基礎(chǔ)結(jié)構(gòu)。
ZSM ArchitectureZSM架構(gòu)ZSM’s perspective is to separate the system into Network Management Domains and E2E Service Management Domain. The domains connect via Domain Integration Fabric. Data Services are cross-domain. Base on the architecture, it further lists the set of services each domain should provide.
ZSM的觀點(diǎn)是將系統(tǒng)分為網(wǎng)絡(luò)管理域和E2E服務(wù)管理域。 域通過域集成結(jié)構(gòu)連接。 數(shù)據(jù)服務(wù)是跨域的。 基于該體系結(jié)構(gòu)，它進(jìn)一步列出了每個(gè)域應(yīng)提供的服務(wù)集。
ETSI GS ZSM 002)ETSI GS ZSM 002 )AS is a higher layer aggregation of those services with closed-loop capability. ETSI is in the process of defining resource closed-loop, network closed-loop, service closed-loop, business closed-loop, and user closed-loop. These overarching closed-loops fulfill the full lifecycle of the inter-layer interaction process, also identifying smaller closed-loop in each resource domain. Each closed-loop is an AS.
AS是具有閉環(huán)功能的那些服務(wù)的高層聚合。 ETSI正在定義資源閉環(huán)，網(wǎng)絡(luò)閉環(huán)，服務(wù)閉環(huán)，業(yè)務(wù)閉環(huán)和用戶閉環(huán)。 這些總體閉環(huán)完成了層間交互過程的整個(gè)生命周期，還識(shí)別了每個(gè)資源域中較小的閉環(huán)。 每個(gè)閉環(huán)都是一個(gè)AS。
Meanwhile, vendors also defined/productized various closed-loop components. For example, Juniper’s “Health Bot”
同時(shí)，供應(yīng)商還定義/生產(chǎn)了各種閉環(huán)組件。 例如，瞻博網(wǎng)絡(luò)的“健康機(jī)器人”
Juniper HealthBot Closed-loop Automation Workflow (Juniper, 2020)瞻博網(wǎng)絡(luò)HealthBot閉環(huán)自動(dòng)化工作流程(2020年6月)“看，我免提！” (“Look, I am hands-free!”)
A scary moment, isn’t it? How to earn the trust of the human operator to let go of the control to ADN? We can get some insights from an example.
可怕的時(shí)刻，不是嗎？ 如何贏得操作員對(duì)ADN放開控制權(quán)的信任？ 我們可以從一個(gè)示例中獲得一些見解。
Since 2018, Google has implemented a fully autonomous data center cooling system to improve operational efficiency and save cooling costs. The system works like this: every five minutes, the cloud-based AI pulls a snapshot of the data center cooling system from thousands of sensors. The AI system feeds the data into its deep neural networks algorithm, which predicts how different combinations of potential actions will affect future energy consumption. The AI system then identifies which actions will minimize energy consumption while satisfying a robust set of safety constraints with a re-enforcement algorithm. Those actions are sent back to the data center, where the suggested actions are verified by the local control system and then implemented. The system delivers an impressive consistent 30% energy saving, which translates to lots of dollars saving at Google scale over a long period. (Yevgeniy S., 2018)
自2018年以來，Google實(shí)施了完全自主的數(shù)據(jù)中心散熱系統(tǒng)，以提高運(yùn)營(yíng)效率并節(jié)省散熱成本。 該系統(tǒng)的工作方式如下：每五分鐘，基于云的AI從數(shù)千個(gè)傳感器中提取數(shù)據(jù)中心冷卻系統(tǒng)的快照。 AI系統(tǒng)將數(shù)據(jù)輸入其深層神經(jīng)網(wǎng)絡(luò)算法，該算法預(yù)測(cè)潛在動(dòng)作的不同組合將如何影響未來的能源消耗。 然后，AI系統(tǒng)識(shí)別出哪些動(dòng)作將最大程度地減少能耗，同時(shí)通過強(qiáng)化算法滿足一組強(qiáng)大的安全約束。 這些操作將被發(fā)送回?cái)?shù)據(jù)中心，在數(shù)據(jù)中心，建議的操作將由本地控制系統(tǒng)進(jìn)行驗(yàn)證，然后予以實(shí)施。 該系統(tǒng)可實(shí)現(xiàn)令人印象深刻的一致30％的節(jié)能，這意味著在Google范圍內(nèi)可以長(zhǎng)期節(jié)省大量資金。 (Yevgeniy S.，2018)
It is interesting to observe how human operators gradually give control to the autonomous system in the process.
有趣的是，觀察人員在此過程中如何逐步將控制權(quán)交給自治系統(tǒng)。
It started as an AI-based recommendation system; human operators took the suggestions and implemented them. This semi-automatic method delivered energy saving but also missed many power-saving opportunities. Because AI suggests a lot of fine-tuning operations to take advantage of the smaller changes in the environment, human operators can not afford to tune the system on all the granular actions at high frequency. The need for autonomous control arose. In 2018, Google completed the building of the closed-loop autonomous control system.
它最初是基于AI的推薦系統(tǒng)； 操作員采納了建議并予以實(shí)施。 這種半自動(dòng)方法可以節(jié)省能源，但也錯(cuò)過了許多節(jié)電機(jī)會(huì)。 由于AI建議進(jìn)行許多微調(diào)操作以利用環(huán)境中較小的變化，因此人工操作員無法承受高頻下所有細(xì)粒度動(dòng)作對(duì)系統(tǒng)進(jìn)行的微調(diào)。 出現(xiàn)了對(duì)自主控制的需求。 Google在2018年完成了閉環(huán)自主控制系統(tǒng)的建設(shè)。
Google took a very cautious approach when introducing the autonomous system into the real world. First, they implemented a set of guardrails. Then they allowed the autonomous system to only control a small range of auto-tuning. Step by step, let the system gradually take over whole control. As described by Joe Kava, Google’s VP of data centers, “And you start to put in the guardrails to make sure that bad things can’t happen, and then you start to launch fully automated systems instead of semiautomated systems”(Yevgeniy S.,2018).
在將自治系統(tǒng)引入現(xiàn)實(shí)世界時(shí)，Google采取了非常謹(jǐn)慎的方法。 首先，他們實(shí)施了一系列的護(hù)欄。 然后，他們?cè)试S自治系統(tǒng)僅控制小范圍的自動(dòng)調(diào)整。 逐步，讓系統(tǒng)逐步接管整個(gè)控制。 正如Google數(shù)據(jù)中心副總裁Joe Kava所描述的那樣，“然后您開始置入護(hù)欄，以確保不會(huì)發(fā)生壞事，然后開始啟動(dòng)全自動(dòng)系統(tǒng)，而不是半自動(dòng)化系統(tǒng)”(Yevgeniy S. ，2018)。
The below chart shows the safety guardrails Google had implemented.
下圖顯示了Google已實(shí)施的安全護(hù)欄。
Google Autonomous Driving DataCenter Cooling System Safeguard Functions (Amanda G., et al., 2018)Google自動(dòng)駕駛數(shù)據(jù)中心冷卻系統(tǒng)保障功能(Amanda G.，et al。，2018)The take away from Google’s experience is that every AS introduced in ADN evolution needs to put safety and robustness as the highest priority concern. We need to know the effect when things go wrong and have a mechanism to isolate the impact and able to put the network back to a known state. Google’s autonomous driving data center cooling control system sets a good example.
從Google的經(jīng)驗(yàn)中脫穎而出的是，在ADN演進(jìn)中引入的每個(gè)AS都需要將安全性和魯棒性作為最高優(yōu)先級(jí)。 我們需要知道出現(xiàn)問題時(shí)的影響，并且需要一種機(jī)制來隔離影響并能夠使網(wǎng)絡(luò)恢復(fù)到已知狀態(tài)。 Google的自動(dòng)駕駛數(shù)據(jù)中心冷卻控制系統(tǒng)就是一個(gè)很好的例子。
Handing over control to AS is scary, and we should be scared. Critical infrastructure, like the network, has a massive impact on society if it goes down, not to mention monetary loss due to SLA violations. I believe it is a better strategy to hand over the control to AS gradually with fallback thought out on each step, like how Telsa does it in delivering the autonomous driving experience.
將控制權(quán)移交給AS令人恐懼，我們應(yīng)該感到恐懼。 諸如網(wǎng)絡(luò)之類的關(guān)鍵基礎(chǔ)架構(gòu)一旦崩潰，將對(duì)社會(huì)產(chǎn)生巨大影響，更不用說由于違反SLA而造成的金錢損失。 我認(rèn)為將控制權(quán)逐步移交給AS是一種更好的策略，并在每個(gè)步驟上都考慮到后備問題，例如Telsa在提供自動(dòng)駕駛體驗(yàn)方面的做法。
The reason is straightforward, as being pointed out by Jeff Mogul of Google (Jeff M. 2018) regarding potential pitfalls of ADN:
正如Google的Jeff Mogul(Jeff M.2018)指出的那樣，原因很簡(jiǎn)單，原因在于ADN的潛在陷阱：
- Any control system (human or automated) has its limits, 任何控制系統(tǒng)(人工或自動(dòng)化)都有其局限性，
- Pushing a control system past its limits can cause crashes, 將控制系統(tǒng)推到極限之外可能會(huì)導(dǎo)致崩潰，
- Sometimes, the problem is not in the control system! 有時(shí)，問題不在控制系統(tǒng)中！
Jeff also argued that unlike autonomous driving vehicles where there is a complete stable manual-driven vehicle as its foundation. In networking, we do not have a perfectly stable manual operated network as our start point for ADN. We need to clean those unstable elements in the system before ADN can take off. “SelfDN success will depend on ‘fixing the environment,’ just as much as on great ML.” Jeff said.
杰夫還認(rèn)為，與自動(dòng)駕駛汽車不同的是，自動(dòng)駕駛汽車具有完全穩(wěn)定的手動(dòng)駕駛汽車作為基礎(chǔ)。 在網(wǎng)絡(luò)中，我們沒有完美穩(wěn)定的手動(dòng)網(wǎng)絡(luò)作為ADN的起點(diǎn)。 我們需要先清理系統(tǒng)中的那些不穩(wěn)定元件，然后才能起飛。 “ SelfDN的成功將取決于'修復(fù)環(huán)境'，以及出色的ML?！?杰夫說。
ADN作為社會(huì)技術(shù)系統(tǒng) (ADN as a Socio-Techincal System)
Technology advancement dramatically increases the feasibility of meeting the technical challenges of an autonomous driving network. But if we just focus on meeting the technical requirements and ignore human, social, and organizational aspects, the ADN can be technically successful but operationally a failure and get rejected by society.
技術(shù)進(jìn)步極大地提高了應(yīng)對(duì)自動(dòng)駕駛網(wǎng)絡(luò)技術(shù)挑戰(zhàn)的可行性。 但是，如果我們只專注于滿足技術(shù)要求，而忽略人員，社會(huì)和組織方面，則ADN在技術(shù)上可能是成功的，但在操作上可能是失敗的，并被社會(huì)所拒絕。
There is a need for a pragmatic approach to engineering ADN as socio-technical systems based on the gradual introduction of socio-technical considerations into existing hardware, software, human-machine interaction development processes.
在將社會(huì)技術(shù)考慮因素逐步引入現(xiàn)有硬件，軟件，人機(jī)交互開發(fā)過程的基礎(chǔ)上，需要一種實(shí)用的方法來將ADN作為社會(huì)技術(shù)系統(tǒng)進(jìn)行工程設(shè)計(jì)。
In 2005, Playchess.com hosted a chess tournament in which teams of human players could use computer assistance during matches. The chess supercomputer Hydra was also entered into the competition. After recently defeating Grand Master Michael Adams 5 ?–? in a six-game match, it was considered the prohibitive favorite. Surprisingly, Hydra was eliminated before the semi-finals, with three of the four semi-finalists consisting of Grand Master-led teams equipped with supercomputers. Even more surprising was the fourth semi-finalist and eventual winner, team ZachS, composed of two relatively amateur chess players named Steven Crampton and Zackary Stephen, using ordinary computers.(Kyle B, John F, 2016)
在2005年，Playchess.com舉辦了一場(chǎng)國(guó)際象棋比賽，在該比賽中，人類運(yùn)動(dòng)員可以在比賽中使用計(jì)算機(jī)協(xié)助。 國(guó)際象棋超級(jí)計(jì)算機(jī)Hydra也參加了比賽。 在最近的六場(chǎng)比賽中以5 ?–?擊敗大師邁克爾·亞當(dāng)斯(Michael Adams)之后，這被認(rèn)為是令人望而卻步的。 令人驚訝的是，九頭蛇在半決賽之前被淘汰，四支半決賽中有三支由大師級(jí)領(lǐng)導(dǎo)的團(tuán)隊(duì)組成，這些團(tuán)隊(duì)配備了超級(jí)計(jì)算機(jī)。 更令人驚訝的是，第四名準(zhǔn)決賽者并最終獲得冠軍的ZachS團(tuán)隊(duì)由兩個(gè)相對(duì)業(yè)余的國(guó)際象棋手史蒂芬·克蘭普頓(Steven Crampton)和扎克里·斯蒂芬(Zackary Stephen)組成，他們使用普通計(jì)算機(jī)(凱爾·B，約翰·F，2016年)
The higher skill level of Hydra and the Grand Masters equipped with supercomputers was not enough to overcome the seamless collaboration between the less skilled amateurs and their weaker computers. As Garry Kasparov stated, “Weak human + machine + better process was superior to a strong computer alone and, more remarkably, superior to a strong human + machine +inferior process.” (Kyle B, John F, 2016)
九頭蛇和配備超級(jí)計(jì)算機(jī)的大師級(jí)別的較高技能水平還不足以克服技術(shù)水平較低的業(yè)余愛好者和較弱的計(jì)算機(jī)之間的無縫協(xié)作。 正如Garry Kasparov所說：“弱人機(jī)+更好的過程優(yōu)于僅強(qiáng)大的計(jì)算機(jī)，更值得注意的是，優(yōu)于強(qiáng)人機(jī)+劣等過程?！?(Kyle B，John F，2016)
The take away from the chess story is that the human-machine combination has the potential to outperform human-alone and computer-alone. The same facts happen in many domains. For example, human forecasters at the National Weather Service can improve the accuracy of computer precipitation forecasts by 25% and computer temperature forecasts by 10% over computer-only predictions, and human-computer teams have the potential to outperform both doctors and computer algorithms at correctly interpreting mammograms. Human is good at spot outliner and can sense certain information that the machine does not have.
從國(guó)際象棋故事中脫穎而出的是，人機(jī)結(jié)合有潛力勝過單人和單機(jī)。 相同的事實(shí)在許多領(lǐng)域中也會(huì)發(fā)生。 例如，美國(guó)國(guó)家氣象局的人類預(yù)報(bào)員可以將計(jì)算機(jī)降水預(yù)報(bào)的準(zhǔn)確性提高25％，將計(jì)算機(jī)溫度預(yù)報(bào)的準(zhǔn)確性提高10％，而人機(jī)團(tuán)隊(duì)則有可能超越醫(yī)生和計(jì)算機(jī)算法。正確解釋乳房X光照片。 人類擅長(zhǎng)于輪廓描繪器，并且可以感知機(jī)器沒有的某些信息。
Socio-technical system design thinking is that the overall design objective is to optimize the system’s collective output efficiency, including organization, people, and technology.
社會(huì)技術(shù)系統(tǒng)設(shè)計(jì)思想是總體設(shè)計(jì)目標(biāo)是優(yōu)化系統(tǒng)的整體輸出效率，包括組織，人員和技術(shù)。
Socio-technical System社會(huì)技術(shù)系統(tǒng)While we are marching to ADN, socio-technical thinking, theory, and design methodologies are invaluable to ensure ADN a success.
當(dāng)我們邁向ADN時(shí)，社交技術(shù)的思想，理論和設(shè)計(jì)方法對(duì)于確保ADN取得成功至關(guān)重要。
As an example, thinking about the future operating model of NOC/SOC (Network Operation Center/Service Operation Center), where is a prominent place that human operators and technologies interact in networking.
例如，考慮一下NOC / SOC(網(wǎng)絡(luò)運(yùn)營(yíng)中心/服務(wù)運(yùn)營(yíng)中心)的未來運(yùn)營(yíng)模式，其中人為運(yùn)營(yíng)商和技術(shù)在網(wǎng)絡(luò)中的交互作用最為突出。
In his article, “5 Overlooked Principles in the Race for Autonomous Networks”, contributed to TMF Insight, Yuval Stein from TEOCO wrote:
TEOCO的Yuval Stein在他的文章“自治網(wǎng)絡(luò)競(jìng)賽中的5個(gè)被忽視的原則”中為TMF Insight做出了貢獻(xiàn)：
“ So long as controllers sit in the NOC/SOC, they will need to understand what’s happening in the network — even in areas where actions are occurring automatically. This means that, at least for the foreseeable future, the NOC will manage network problems with a mixture of manual and automatic resolutions. The needs of the NOC/SOC should continue to be considered — and respected.”
只要控制器位于NOC / SOC中，他們就需要了解網(wǎng)絡(luò)中正在發(fā)生的事情，即使是在自動(dòng)執(zhí)行操作的區(qū)域中也是如此。 這意味著，至少在可預(yù)見的將來，NOC將通過手動(dòng)和自動(dòng)解決方案的組合來管理網(wǎng)絡(luò)問題。 應(yīng)該繼續(xù)考慮并尊重NOC / SOC的需求。 ”
He further elaborated:
他進(jìn)一步闡述：
“NOC/SOC systems need to:
“ NOC / SOC系統(tǒng)需要：
Show relevant alarms at all layers. As noted, modern management systems are able to filter a large percentage of symptomatic alarms, but existing problems need to be listed.
在所有層顯示相關(guān)警報(bào)。 如前所述，現(xiàn)代管理系統(tǒng)能夠過濾大部分癥狀警報(bào)，但是需要列出現(xiàn)有問題。
Root-cause analysis is still required, even if the problem is resolved by a lower-layer management system. The NOC needs to have a clear, holistic view to understand what was resolved automatically — and what needs manual intervention.
即使通過較低層的管理系統(tǒng)解決了問題，仍然需要進(jìn)行根本原因分析。 NOC需要有清晰的整體視圖，以了解自動(dòng)解決的問題以及需要手動(dòng)干預(yù)的問題。
Enrich alarms with organizational data that assists controllers. Information like geography, administrative ownership, and network relevant change requests must be associated to alarms.
利用有助于控制者的組織數(shù)據(jù)來豐富警報(bào)。 諸如地理位置，管理所有權(quán)和與網(wǎng)絡(luò)相關(guān)的變更請(qǐng)求之類的信息必須與警報(bào)相關(guān)聯(lián)。
Track the source of abnormalities, whether they are alarm-based or measurement-based. NOC controllers should understand the nature of the abnormalities as much as possible, which may require additional alarms, measurements or metadata.”
跟蹤異常源，無論是基于警報(bào)還是基于測(cè)量。 NOC控制器應(yīng)盡可能了解異常的性質(zhì)，這可能需要其他警報(bào)，測(cè)量或元數(shù)據(jù)。 ”
This kind of insight into the human operator’s needs when interacting with technology in future ADN is critical.
在將來的ADN中與技術(shù)交互時(shí)，這種對(duì)操作員需求的洞察至關(guān)重要。
摘要 (Summary)
Inspired by the rapid development of autonomous driving vehicles in recent years, ADN is the most ambitious goal in the networking industry ever. The formula of AI+Software Defined Network makes this goal seemly achievable. Vendors and standard organizations chart out the 6 phases of development, from mostly manual operation L0 to L5 of fully autonomous operation. The race to ADN already started, vendors not only publish white papers but also align their product strategies accordingly. Standard organizations also speed up the process of identifying requirements and defining the reference architectures.
受近年來自動(dòng)駕駛汽車快速發(fā)展的啟發(fā)，ADN是網(wǎng)絡(luò)行業(yè)有史以來最雄心勃勃的目標(biāo)。 AI +軟件定義網(wǎng)絡(luò)的公式使這一目標(biāo)似乎可以實(shí)現(xiàn)。 供應(yīng)商和標(biāo)準(zhǔn)組織列出了六個(gè)開發(fā)階段，從手動(dòng)操作L0到完全自主操作的L5。 與ADN的爭(zhēng)奪已經(jīng)開始，供應(yīng)商不僅發(fā)布白皮書，而且相應(yīng)地調(diào)整其產(chǎn)品策略。 標(biāo)準(zhǔn)組織還加快了確定需求和定義參考體系結(jié)構(gòu)的過程。
Our current network infrastructure has stability issues that need to be addressed; in other words, cleaned up to set a good foundation for ADN.When designing ADN, we need to think beyond AI and networking technology; the socio-technical approach is essential.
我們當(dāng)前的網(wǎng)絡(luò)基礎(chǔ)架構(gòu)存在穩(wěn)定性問題，需要解決。 在設(shè)計(jì)ADN時(shí)，我們需要思考的不僅僅是AI和網(wǎng)絡(luò)技術(shù)； 社會(huì)技術(shù)方法至關(guān)重要。
DARPA’s 2004 $1 million Grand Challenge for Autonomous Vehicle kicked off the rapid development of Autonomous Vehicle. Should we have such an event to kick start Autonomous Driving Network? K. Kompella, Juniper CTO, definitely thought so. He proposed the below “The Networking Grand Challenge”.
DARPA 2004年的100萬(wàn)美元的無人駕駛汽車大挑戰(zhàn)賽開始了無人駕駛汽車的飛速發(fā)展。 我們是否應(yīng)該有這樣的活動(dòng)來啟動(dòng)自動(dòng)駕駛網(wǎng)絡(luò)？ 瞻博網(wǎng)絡(luò)首席技術(shù)官K. Kompella絕對(duì)是這樣認(rèn)為的。 他提出了以下“網(wǎng)絡(luò)大挑戰(zhàn)”。
Self-Driving Network Grand Challenge (K. Kompella, Juniper 2019)無人駕駛網(wǎng)絡(luò)大挑戰(zhàn)(K.Kompella，瞻博網(wǎng)絡(luò)2019)Of course, we all know that telecom network like other infrastructure systems evolves over a long time and not have many opportunities to be built “from scratch”; instead the new or changed elements are always fitted into the previously made infrastructure. But this “clean-slate” exercise does have lots of value for validating our assumption, identifying the gaps and potential pitfalls in technology, and developing the new algorithms. Potentially this can spark some disruptive new technology. So I also think that is a good idea worth exploring.
當(dāng)然，我們都知道，電信網(wǎng)絡(luò)像其他基礎(chǔ)設(shè)施系統(tǒng)一樣，是經(jīng)過很長(zhǎng)時(shí)間發(fā)展的，并且沒有很多“從頭開始”建立的機(jī)會(huì)。 取而代之的是，總是將新的或更改的元素安裝到先前制作的基礎(chǔ)結(jié)構(gòu)中。 但是，這種“干凈的”練習(xí)對(duì)于驗(yàn)證我們的假設(shè)，確定技術(shù)差距和潛在陷阱以及開發(fā)新算法確實(shí)具有很多價(jià)值。 這有可能引發(fā)一些破壞性的新技術(shù)。 因此，我也認(rèn)為這是一個(gè)值得探討的好主意。
Mark W. (2014) Where to a history of autonomous vehicles.
馬克·W(Mark W.)(2014) 自動(dòng)駕駛汽車的歷史 。
Nick F, Jennifer R (2017) Why (and How) Networks Should Run Themselves
Nick F，Jennifer R(2017 )網(wǎng)絡(luò)為什么(以及如何)自己運(yùn)行
K. Kompella (2017) The Self-Driving Network: How to Realize It
K.Kompella(2017) 自駕車網(wǎng)絡(luò)：如何實(shí)現(xiàn)
K. Kompella (2018) The Self Driving Network: from vision to execution
K.Kompella(2018) 自我駕駛網(wǎng)絡(luò)：從愿景到執(zhí)行
K. Kompella (2019) Self Driving Networks: Looks Mom, No Hands
K.Kompella(2019) 自我駕駛網(wǎng)絡(luò)：看起來媽媽，沒有手
Juniper (2020) Health Bot Overview
瞻博網(wǎng)絡(luò)(2020) 衛(wèi)生機(jī)器人概述
Ciena (2019) Introducing the Adaptive Network Vision
Ciena(2019) 介紹自適應(yīng)網(wǎng)絡(luò)愿景
TMF (2019) Autonomous Networks: Empowering Digital Transformation For the Telecoms Industry
TMF(2019) 自主網(wǎng)絡(luò)：助力電信行業(yè)的數(shù)字化轉(zhuǎn)型
Huawei (2018) Huawei’s David Wang: Moving Towards Autonomous Driving Networks
華為(2018) 華為的David Wang：走向自動(dòng)駕駛網(wǎng)絡(luò)
Huawei (2020) Huawei ADN Solution White Paper
華為(2020) 華為ADN解決方案白皮書
Cisco (2019) Digital Network Readiness Model
思科(2019) 數(shù)字網(wǎng)絡(luò)就緒模型
David W. (2018) Moving toward Autonomous Driving Network
David W.(2018) 向自動(dòng)駕駛網(wǎng)絡(luò)邁進(jìn)
Huawei (2020) Huawei ADN Solution White Paper
華為(2020) 華為ADN解決方案白皮書
Cisco (2019) Cisco Digital Network Readiness Model
思科(2019) 思科數(shù)字網(wǎng)絡(luò)就緒模型
CRN (2019) Juniper Networks CEO: ‘The Goal Now Is A Self-Driving Network’
CRN(2019) 瞻博網(wǎng)絡(luò)首席執(zhí)行官：``現(xiàn)在的目標(biāo)是自駕車網(wǎng)絡(luò)''
Analysys Mason (2018) From Autonomous to Adaptive: The Next Evolution in Networking
Analysys Mason(2018) 從自主到自適應(yīng)：網(wǎng)絡(luò)的下一個(gè)發(fā)展
Joseph S. (2019) Autonomous Systems — A Rigorous Architectural Characterization
Joseph S.(2019) 自治系統(tǒng)-嚴(yán)格的建筑特征
Timo P. (2018) The future of network and service automation
Timo P.(2018) 網(wǎng)絡(luò)和服務(wù)自動(dòng)化的未來
Zeng W., et al.,(2017) Big data in telecommunication operators: data, platform and practices
Zeng W.，et al。，(2017) 電信運(yùn)營(yíng)商中的大數(shù)據(jù)：數(shù)據(jù)，平臺(tái)和實(shí)踐
Wylie W. (2017) Juniper Says its Bots are a Step Toward a “Self-Driving Network”
Wylie W.(2017) 瞻博網(wǎng)絡(luò)表示，其機(jī)器人是邁向“自動(dòng)駕駛網(wǎng)絡(luò)”的一步
ETSI (2019) ZSM requirements ETSI GS ZSM 001
ETSI(2019)ZSM要求ETSI GS ZSM 001
ETSI (2019) ZSM reference architecture ETSI GS ZSM 002
ETSI(2019)ZSM參考架構(gòu)ETSI GS ZSM 002
Yevgeniy S. (2018) Google is Switching to a Self-Driving Data Center Management System
Yevgeniy S.(2018) Google正在切換到自動(dòng)駕駛數(shù)據(jù)中心管理系統(tǒng)
Amanda G., et al., (2018) Safety-first AI for autonomous data center cooling and industrial control.
Amanda G.，et al。，(2018) 安全第一的AI，用于自主數(shù)據(jù)中心散熱和工業(yè)控制。
Jeff M. Unsafe at Any Speed: Self-Driving Network without Shelf-Crushing
杰夫·M(Jeff M.) ：任何速度都不安全：無人駕駛的自動(dòng)駕駛網(wǎng)絡(luò)
Yuval S. 5 overlooked principles in the race for autonomous networks
Yuval S.5 在自治網(wǎng)絡(luò)競(jìng)賽中被忽略的原則
Nils Nillson, 1996 Introduction of Teleo-Reactive Programs
Nils Nillson，1996年，采用Teleo-Reactive程序
Kyle B, John F, 2016 Autonomous Systems to Sociotechnical Systems: Designing Effective Collaborations
Kyle B，John F，2016年從自治系統(tǒng)到社會(huì)技術(shù)系統(tǒng)：設(shè)計(jì)有效的協(xié)作
翻譯自: https://medium.com/swlh/autonomous-driving-network-adn-on-its-way-772d053f8a1e
自動(dòng)駕駛發(fā)展
總結(jié)
以上是生活随笔為你收集整理的自动驾驶发展_自动驾驶网络及其发展的全部?jī)?nèi)容，希望文章能夠幫你解決所遇到的問題。
- 上一篇： 诺贝尔奖距离中国内地多远(ZT)
- 下一篇： Axure交互之链接行为