文章結構: Abstract Categories and Subject Descriptors General Terms Keywords 1. INTRODUCTION 2. CONTROL & MANAGEMENT TODAY 2.1 Reachability Control in Enterprises 2.2 Peering Policies in Transit Networks 2.3 Same Problems, Many Guises 2.4 Moving Forward 3. THE 4D ARCHITECTURE 3.1 Design Principles 3.2 New 4D Network Architecture 3.3 Advantages of the 4D Architecture 3.4 Challenges for the 4D Architecture 3.5 Routing Control Platform 4. RESEARCH AGENDA 4.1 Decision Plane 4.1.1 Algorithms Satisfying Network Level Objectives 4.1.2 Coordination Between Decision Elements 4.1.3 Introducing Hierarchy in the Decision Plane 4.2 Dissemination Plane 4.3 Discovery Plane 4.4 Data Plane 5. EVALUATING NEW ARCHITECTURES 5.1 Experimental Platforms 5.2 Opportunities for Field Deployment 6. RELATED WORK 7. CONCLUDING REMARKS ================================================================================================ Outline: The whole article can be divided into eight parts. The first part is the content before INTRODUCTION, including Abstract, Categories and Subject Descriptors, General Terms, Keywords. The Abstract tell us the disadvantages of the current Internet architecture. In order to conquer the disadvantages, the authors propose the 4D architecture. the second part:INTRODUCTION. This part tells us the IP networking is too old to satisfy new service request. the third part:CONTROL & MANAGEMENT TODAY. this part tells us in real world, some kinds of problems make the networks hard to control, and radically the problems are similar. the forth part:THE 4D ARCHITECTURE. This part tells us the work the authors did, including the design principles of the new architecture ,details and advantages of the new architecture, challenges for the new architecture. the fifth part:RESEARCH AGENDA. This part tells us what we can do or what we will do. This part is divided into 4 parts, Decision Plane, Dissemination Plane, Discovery Plane, Data Plane respectively. the sixth part: EVALUATING NEW ARCHITECTURES. this part tells us how we can evaluate the new architecture. the seventh part: RELATED WORK. the eighth part: CONCLUDING REMARKS. this part concludes the whole article. 每一段落的展開方法: 1:INTRODUCTION (1): deduction (2): deduction (3): deduction (4): induction (5): generalization 2: CONTROL & MANAGEMENT TODAY (1): time (2): deduction (3): deduction (4): deduction (5): induction (6): generalization 2.1: (1): time (2): example (3): example (4): induction (5): generalization 2.2: (1): induction (2): example (3): example (4): induction (5): generalization 2.3: (1): induction (2): induction 2.4: generalization 3: THE 4D ARCHITECTURE (1): induction 3.1: (1): induction (2): definition (3): definition (4): definition (5): generalization 3.2: (1): induction (2): definition (3): definition (4): definition (5): definition (6): generalization 3.3: (1): induction (2): definition (3): definition (4): definition (5): definition (6): definition 3.4: (1): induction (2): definition (3): definition (4): definition (5): definition (6): definition 3.5: (1): induction (2): generalization 4. RESEARCH AGENDA (1): induction (2): induction 4.1: definition 4.1.1: definition 4.1.2: definition 4.1.3: definition 4.2: definition 4.3: definition 4.4: definition 5. EVALUATING NEW ARCHITECTURES (1): induction 5.1 Experimental Platforms (1): induction (2): example (3): example 5.2 Opportunities for Field Deployment (1): induction (2): time (3): generalization (4): generalization 6. RELATED WORK induction 7. CONCLUDING REMARKS induction 全部重要句型; 1: 被動語態 The AS-level objectives are specified in the decision plane, and enforced through direct configuration of the state that drives how the data plane forwards packets. 2: 定語從句 A good example of this principle is the Ethernet technology, which has successfully evolved from a shared-medium network to a switched network with new control-plane protocols based on learning and spanning trees, all while leaving the packet format unchanged. 3: 狀語從句 In this example, the problems arise because the ability of a network to carry packets depends on the routing protocols and the packet filtersworking in concert. 4)全部重要單詞和對應中文; network management 網絡管理 robustness 健壯性,魯棒性 clean slate 零基礎 reachability 可達性 configuration 配置 peering 對等的 autonomous systems 自治系統 architecture 體系結構 decision plane 決策層 dissemination plane分發層 discovery plane 發現層 data plane 數據層 routing control platform 路由控制平臺 scalability 可擴展性 NAT(network address translation) 網絡地址轉換 ××× (virtual private network) 虛擬專用網 5)全部的習慣用法,包括主謂搭配、動賓搭配、形容詞與名詞的搭配、副詞與動詞的搭配、介詞與名詞的搭配,以及短語動詞等等; we present ... that ... in order to handle ... It might be desirable to ... ... is based on ... in addition to ... it has since ... Our notion of ... is ... ... means ... Because of ..., ... ... is responsible for ... ... focus on ... 6)重點研究若干段落中的冠詞用法和時態的用法; INTRODUCTION中的第一段 Although IP networking has been wildly successful, there are serious problems lurking 「under the hood.」 「under the hood.」在引擎蓋下。 IP networks exhibit a defining characteristic of unstable complex systems—a small local event (e.g., misconfiguration of a routing protocol on a single interface) can have severe, global impact in the form of a cascading meltdown. 「a defining characteristic」,一項基本特徵。泛指。 「a small local event」,一個小的局部事件。泛指。 「a routing protocol」,一種路由協議。泛指 「a single interface」,一個單一接口。泛指。 「in the form of」。以...的形式。固定用法。 In addition, individual Autonomous Systems (ASes) must devote significant resources to 「working around」 the constraints imposed by today’s protocols and mechanisms to achieve their goals for traffic engineering, survivability, security, and policy enforcement. 「the constraints」,限定用法。 We believe the root cause of these problems lies in the control plane running on the network elements and the management plane that monitors and configures them. 「the root cause of these problems」,這些問題的根本緣由。限定cause。 「the control plane 」,「the management plane」。表示特定稱謂。 In this paper, we argue for revisiting the division of functionality and advocate an extreme design point that completely separates a network’s decision logic from the the protocols that govern interaction of network elements 在本文中,咱們主張從新劃分功能,倡導一種新的的設計要點,將網絡的決策邏輯與管理網絡元件的協議徹底分離開。 「the protocols that ...」 ...的協議。表示限定。 「a network’s decision logic 」。網絡的決策邏輯。泛指。 We initially focus our attention on the operation of a single Autonomous System (AS), though we also discuss how multiple ASes can coordinate their actions. 」a single Autonomous System 「一個單一的自治系統。泛指。 時態。本文大部分句子用的是通常如今時。當介紹到舊的技術時,用的是過去時。 7)參考引用他人做品的方式; 若是引用的是網頁,則附上網頁中文章的標題和網址。 如:NSF CISE, 「The GENI initiative.」 http://www.nsf.gov/cise/geni/. 若是引用的是論文,則附上做者,文章標題,所發表的期刊或會議。 如:L. Peterson, Y. Gottlieb, M. Hibler, P. Tullmann, J. Lepreau,S. Schwab, H. Dandekar, A. Purtell, and J. Hartman, 「A NodeOS interface for active networks,」 IEEE J. Selected Areas in Communications, March 2001. 8)全文的寫做特色; 文章首先在摘要中爲本身的文章中的觀點作簡要的說明。用如今的網絡體系結構中的弱點引出做者所研究的網絡體系結構的優勢。按部就班地講述本身的研究。用了大量的實例來論證和支撐本身的觀點。用了大量的圖來講明本身想表達的意思,由於使用的大量的圖,因此簡化了說明,有利於讀者理解。本文邏輯嚴謹,結構緊湊。概括演繹綜述得當。 9)從英語寫做的角度出發,回答問題:經過認真的閱讀和研究該文章,你有何收穫? 經過閱讀本文,我體會到英文寫做的邏輯嚴密性。每一部分與每一部分之間,段落與段落之間,句子與句子之間,都有比較強的邏輯關係。結構緊湊。 閱讀的過程當中發現大部分的句子使用的是被動語態。使用被動語態可使說理顯得客觀。 遇到了一些難懂的詞彙和詞組,用Google, StarDict, dict.cnki.net, translate.google.cn 查找合適的釋義。因此在我之後的學習中,要注意詞彙量的積累。 本文平均每句有23個詞。難度爲17。說明本文是偏難的。本文中有不少的句型我可能會在之後的學習和研究中用到,因此應該注意積累。