计算机网络技术的历史和新进展

计算机网络技术的历史和新进展

高速计算机信息网络是信息社会的神经和血管 • 体系结构：网络的骨架和神经 • 协议：网络的心脏和血液

主要内容 • 网络概述 • Internet的发展和成功经验 • 计算机网络技术的历史回顾 • 国际高速计算机网络研究计划 • 中国高速计算机网络研究计划

What is a Network?from end system point of view • Network offers a service: move information • bird, fire, messenger, truck, telegraph, telephone, Internet … • another example, transportation service: move objects • horse, train, truck, airplane ... • What distinguish different types of networks? • The services they provide • What distinguish the services? • latency • bandwidth • loss rate • number of end systems • service interface (how to invoke?) • other details • reliability, unicast vs. multicast, real-time, message vs. byte ...

What is a Network?Infrastructure Centric View • Electrons and photons as communication medium • Links: fiber, copper, satellite, … • Switches: mechanical/electronic/optical, crossbar/Banyan • Protocols: TCP/IP, ATM, MPLS, SONET, Ethernet, PPP, X.25, FrameRelay, AppleTalk, IPX, SNA • Functionalities: routing, error control, congestion control, Quality of Service (QoS) • Applications: FTP, WEB, X windows, ...

Types of Networks • Geographical distance • Local Area Networks (LAN): Ethernet, Token ring, FDDI • Metropolitan Area Networks (MAN): DQDB, SMDS • Wide Area Networks (WAN): X.25, ATM, frame relay • Information type • data networks vs. telecommunication networks • Application type • special purpose networks: airline reservation network, banking network, credit card network, telephony • general purpose network: Internet

Types of Networks • Right to use • private: enterprise networks • public: telephony network, Internet • Ownership of protocols • proprietary: SNA • open: IP • Technologies • terrestrial vs. satellite • wired vs. wireless • Protocols • IP, AppleTalk, SNA

计算机网络发展历史回顾 • 七十年代的计算机网络 • X.25 分组交换网：各国的电信部门建设运行 • 各种专用的网络体系结构：SNA，DNA • Internet 的前身ARPANET进行实验运行 • 八十年代的计算机网络 • 标准化计算机网络体系结构：OSI • 局域网络 LAN 技术空前发展 • 建成NSFNET，Internet 初具规模 • 九十年代的计算机网络 • Internet空前发展 • Web技术在Internet/Intranet 得到广泛应用

The Internet • Global scale, general purpose, heterogeneous-technologies, public, computer network • Internet Protocol • open standard: Internet Engineering Task Force (IETF) as standard body • technical basis for other types of networks • Intranet: enterprise IP network • Developed by the research community

History of the Internet • 70’s: started as a research project, 56 kbps, < 100 computers • 80-83: ARPANET and MILNET split, • 85-86: NSF builds NSFNET as backbone, links 6 Supercomputer centers, 1.5 Mbps, 10,000 computers • 87-90: link regional networks, NSI (NASA), ESNet(DOE), DARTnet, TWBNet (DARPA), 100,000 computers • 90-92: NSFNET moves to 45 Mbps, 16 mid-level networks • 94: NSF backbone dismantled, multiple private backbones • Today: backbones run at 2.5 / 10 Gbps, 10s millions computers in 150 countries

Growth of the Internet • Number of Hosts on the Internet: Aug. 1981 213 Oct. 1984 1,024 Dec. 1987 28,174 Oct. 1990 313,000 Oct. 1993 2,056,000 Apr. 1995 5,706,000 Jul. 1997 19,540,000 Jul. 2000 93,047,785

Recent Growth (1991-2000)

Internet 发展规模和趋势 • Internet的发展速度 • 是历史上发展最快的一种技术 • 以商业化后达到 5000 万用户为例 • 电视用了13年，收音机用了38年，电话更长 • Internet 从商业化后达到 5000 万用户用了4 年时间 • Internet 正在以超过摩尔定理的速度发展

网络时代的三大基本定律 摩尔定律: CPU性能18个月翻番,10年100倍。所有电子系统（包括电子通信系统，计算机）都适用光纤定律：超摩尔定律，骨干网带宽9个月翻番，10年10000倍。带宽需求呈超高速增长的趋势迈特卡尔夫定律:联网定律，网络价值随用户数平方成正比。未联网设备增加N倍，效率增加N倍。联网设备增加N倍，效率增加N2倍

网络带宽与CPU性能

光纤容量

容量光纤长度特点研制单位 3Tb/s 40km 用 T-EDFA NTT (160Gb/s X 19 CH OTDM/WDM) (DSF-0=1535nm) 用DSF光纤(NZDSF) PD1-1 ----------------------------------------------------------------------------------------------------------------------------------------- 1.02Tb/s 1000km 0.4b/s/Hz，用 SMF光纤 CNET (20Gb/s x 51 CH WDM) (SMF 环测) 101km放大器间距 PD4-1 ----------------------------------------------------------------------------------------------------------------------------------------- 1Tb/s 342km 用True Wave光纤Lucent (40Gb/s x 25 CH) (True Wave 光纤) 85 km放大器间距PD7-1 ----------------------------------------------------------------------------------------------------------------------------------------- 750Gb/s 2000km 采用 C波段和 L波段 Tyco (5.3Gb/s x 50 CH,10Gb/s x 8CH) (纯Si光纤+NZDSF环测) 纯Si光纤 PD16-1 ----------------------------------------------------------------------------------------------------------------------------------------- 640Gb/s 7200km 0.33b/s/Hz Tyco (10Gb/s x 64 CH) (NZDSF 环测) PD2-1 ----------------------------------------------------------------------------------------------------------------------------------------- 490Gb/s 335.2KM 采用拉曼放大+EDFA混合 Lucent (10Gb/s x 49 CH) (ZDSF 环测 ) 0色散光纤(67km)+NZDSF(17km)PD8-1 ----------------------------------------------------------------------------------------------------------------------------------------- 340Gb/s 6380km 实线试验 Alcatel (10Gb/s x 34 CH WDM) (NZDSF) 50GHz Spacing PD18-1 ----------------------------------------------------------------------------------------------------------------------------------------- 80Gb/s 172km Soliton Chalmer (10Gb/s x 8 CH OTDM) (DSF 0=1547nm) 已敷设的光纤 PD6-1 高水平大容量光纤传输试验系统

180 160 140 120 100 80 60 40 20 1996 1997 1998 1999 2000 2001 2002 Data (Still) Overtaking Voice International data traffic already exceeds international voice from Australia and Scandinavia. Voice RelativeCapacity(%) Data 0 Source: MCI (Vint Cerf)

IP (Still) Conquering Data Traffic Ratios 100% 90% All Other 80% IPX 70% TCP/IP 60% 50% 40% 30% 20% 10% 0% 1998 2000 1996 Source: Gartner 1997

Multimedia Dynamic WWW Static WWW FTP and Telnet E-Mail and News Other Internet (Still) Going Interactive To Transactional Pages (Red) andAudio/Video Content (Purple) 2% 100% 7% 7% 14% 16% 80% 27% Relative User Population 28% 23% 60% 15% 40% 39% 23% 12% 20% 17% 18% 17% 14% 13% 0% 8% 1998 2000 1996 Source: The Yankee Group, 1996

Apps (Always) Driving Capacity 155 ATM/POS Virtual Reality, Medical Imaging 3 T3/E3 Video Conferencing, MPEG1 NTSC Video 1.5 T1/E1 Simple Video, Multimedia .128 ISDN, Frame Relay Browsing, PCM Voice .0288 New Modem IP, PCS, E-Mail, File Transfer .0192 Old Modem Telnet, VoIP .004 Wireless WAN Paging Mb/s Minimum Bandwidth for Application per User

Internet Data/Voice/Video Transport Convergence

高速信息网络的发展方向：通信与计算聚合 • 通信和计算技术的聚合 • 改变了各自的原有特征 • 高速信息网络体系结构的发展趋势 • 分层结构；分布控制、管理和安全机制 • 分层结构 • 比特路层 • 服务层 • 应用层

高速信息网络的体系结构 • 比特路层 • 主干网传输技术：SDH/SONET，光纤 • 主干网交换技术：IP over SDH或光纤，GbE，支持IPv6 • 端系统接入技术：LAN；ADSL、FTTH、HFC • 服务层（支撑技术） • 全球统一的地址、域名；安全的系统管理和访问控制 • Browser/Server 计算模式，支持 Data, Voice, Video • 以 Java 为代表的网络编程语言 • 应用层（用户功能） • 用户－用户（立即响应，可适当延迟） • 用户－服务器（立即响应，可适当延迟）

IP Transport Alternatives IP over ATM IP over SDH/SONET IP over Optical B-ISDN Multiplexing, Protection, and Management at Every Layer. IP Long-Term Winners ATM IP IP SDH/SONET IP SDH/SONET ATM Optical Optical Optical Optical Eliminating Layers Lowers Costs.

Who is Who on the Internet ? • Internet Engineering Task Force (IETF):The IETF is the protocol engineering and development arm of the Internet. Subdivided into many working groups, which specify Request For Comments or RFCs. • IRTF (Internet Research Task Force):The Internet Research Task Force is a composed of a number of focused, long-term and small Research Groups. • Internet Architecture Board (IAB): The IAB is responsible for defining the overall architecture of the Internet, providing guidance and broad direction to the IETF. • The Internet Engineering Steering Group (IESG): The IESG is responsible for technical management of IETF activities and the Internet standards process. Standards. Composed of the Area Directors of the IETF working groups.

Internet Standardization Process • All standards of the Internet are published as RFC (Request for Comments). But not all RFCs are Internet Standards ! • available: http://www.ietf.org • A typical (but not only) way of standardization is: • Internet Drafts • RFC • Proposed Standard • Draft Standard (requires 2 working implementation) • Internet Standard (declared by IAB) • David Clark, MIT, 1992: "We reject: kings, presidents, and voting. We believe in: rough consensus and running code.”

Services Provided by the Internet • Shared access to computing resources • telnet (1970’s) • Shared access to data/files • FTP, NFS, AFS (1980’s) • Communication medium over which people interact • email (1980’s), on-line chat rooms, instant messaging (1990’s) • audio, video (1990’s) • replacing telephone network? • A medium for information dissemination • USENET (1980’s) • WWW (1990’s) • replacing newspaper, magazine? • audio, video (1990’s) • replacing radio, CD, TV?

Today’s Vision • Everything is digital: voice, video, music, pictures • Everything is on-line: bank statement, medical record, books, airline schedule, weather, highway traffic, toaster, refrigerator … • Everyone is connected: doctor, teacher, broker, mother, son, friends, enemies

What is Next? • Electronic commerce • virtual enterprise • Internet entertainment • interactive sitcom • World as a small village • community organized according to interests • enhanced understanding among diverse groups • Electronic democracy • little people can voice their opinions to the whole world • bridge the gap between information haves and have no’s • Electronic terrorism • hacker can bring the whole world to its knee

Industrial Players • Telephone companies • own long-haul and access communication links, customers • Cable companies • own access links • Wireless/Satellite companies • alternative communication links • Utility companies: power, water, railway • own right of way to lay down more wires • Medium companies • own content • Internet Service Providers • Equipment companies • switches/routers, chips, optics, computers • Software companies

Internet Physical Infrastructure Backbone ISP • Residential Access • Modem • DSL • Cable modem • Satellite • LAN ISP • Enterprise/ISP access, Backbone transmission • T1/T3, DS-1 DS-3 • OC-3, OC-12 • ATM vs. SONET, vs. WDM • Campus network • Ethernet, ATM • Internet Service Providers • access, regional, backbone • Point of Presence (POP) • Network Access Point (NAP)

Links for Long Haul Transmission • Possibilities • IP over SDH/SONET • IP over ATM • IP over Frame Relay • IP over WDM • Types of links • T1/DS1: 1.544 Mbps • T3/DS3: 44.736 Mbps • STS-1/OC-1: 51.850 Mbps • STS-3/OC-3: 155.2 Mbps • STS-12/OC-12: 622.080 Mbps • STS-48/OC-48: 2.488 Gbps • STS-192/OC-192: 9.953 Gbps • Higher levels of services offered commercially • Frame Relay • ATM

Internet 的成功经验 • 有远见的政府不断支持：1969－ • 有风险的企业参与和投入： • NFS：MCI、IBM • vBNS：MCI；Abilene: Qwest，CISCO • 联合协作的开放式研究：IETF/RFC • 教育和科研的示范网络为起点 • 具有实验物理学的研究特点 • ARPAnet、NSF、ANS、vBNS • 简单实用的技术路线：TCP/IP

Commercialization Privatization 21st Century Interoperable Networking High Performance SprintLink Research &Education InternetMCI Networks US Govt Networks ANS ARPAnet NSFNET Active gigabit Nets testbeds wireless Internet2, Abilene, vBNS WDM Advanced US Govt Networks Quality of Service (QoS) Research and Development Partnerships

Outline • The first design for a packet-switched network by Paul Baran in early 60’s • TCP/IP protocol: one realization of the basic principles in Baran’s design • How the Internet looks like today • Where it may be moving to

DST Distce Nxt-H Destination address data Paul Baran’s classical work“On Distributed Communications” • Time: 1960 - 1964 • Goal: building a robust communication system that could survive nuclear attacks • Outcome: a packet-switched network

What is in Baran’s Design A self-adaptive system: hot-potato routing • If don’t know better: forwarding packets to all neighbors • Update routing table by observing packets passing by; old routing entries timeout and deleted • Forward packets ASAP • not necessarily along shortest paths all the time Learning & adapting to the changing environment

What is in Baran’s Design (cont’) Datagram delivery • Each switch makes packet forwarding decision based on its own routing table • Each packet is forwarded independently from any others • Switches keep no state about end nodes • Not a most efficient network • Delivery will not be perfect End systems must tolerate & recover from transmission errors

What is in Baran’s Design (cont’) A distributed system • all switch nodes are equal • eliminating any single point of failure • components may fail, the system must not • system robustness through • adequate physical redundancy • adaptive routing Simulation experiment demonstrated that “extremely survivable networks can be built using a moderately low redundancy of connectivity level”—Paul Baran, 1964

some less known side of the story... at the time • a number of people in telecommunication industry considered Baran’s proposal “totally preposterous” • “they kicked, screamed, grumbled, and worse. Their response tended to be emotional, often with anger, rarely with humor…” “outsiders could not possibly understand the complexity of large systems like the telephone network”

Different approaches to system reliability • phone system at the time (still true today) • Dumb end nodes, smart network • making each and every network component reliable • system reliability = component reliability • high component reliability by local redundancy • everything expected to work; failures expected to be extreme exceptions • a human configured, tightly controlled system • Baran proposal • a reliable system built out of simple, unreliable parts • an adaptive system that adjusts itself to changes automatically • Smart ends to fix transmission errors

One Realization of Baran’s Desing:the Internet • interconnecting heterogeneous subnets • two basic functionalities: • globally unique addresses • datagram delivery from sources to destinations via dynamic routing Claim: simple, flexible, scalable, and robust all kinds of applications all kinds of transport protocols IP’s view of the world IP All kinds of subnet technologies

Datagrams as the basic building block: System Simplicity • Each packet carries its own address • One routing table serves all traffic • An enabler to the explosive growth • the simpler, the fewer chances to go wrong • the simpler, the easier to grow • the requirement of least common network functionality maximizes the number of usable networks

Datagrams as the basic building block: Flexibility • “runs over anything” • IP started with ARPANET, PRNET, SATNET; all long gone • today IP runs over 100 Mb or Gb/s Ethernet,FDDI, Frame Relay, ATM, SONET, DWDM … • Supports all different applications • used to be telnet, ftp, email only • now audio, video, web, e-commerce, online-education What will the future bring? • The best bet would be to stay flexible

Datagrams as the basic building block: Scalability To scale the system must be able to gracefully handle • the growth in the total number of end systems • the growth in traffic volume • the growth in network size • larger routing tables • More frequent changes With IP, “the network knows nothing about individual end applications; end applications know nothing about network internals”—Van Jacobson • made it possible to aggregate routing table entries according to scaling need

Datagrams as the basic building block: Robustness • self-adaptive nature of dynamic routing facilitated the growth • dynamic routing and datagram delivery go hand-in-hand • periodic routing updates: “are you still there” • silent assumption: existing parts may fail, new parts may appear any time  changes considered the norm rather than exceptions • trades off optimal link usage (e.g. header overhead, update overhead) for system robustness

“Unintended consequence” ? In a casual conversation Noel Chiappa said, [without DoD’s Internet research initiative] “Perhaps we’d have discovered computer networks eventually, but I have no idea when; ... It’s also unclear if we’d have the kind of network we do if it hadn’t been intended for such a use. “A spirit of intense robustness, and the ability to keep going no matter what, has been part of the ‘Internet ethic’ since day one, and it is due in part to the rather severe operation environment for which it was originally intended. This extremely monolithic nature of the Internet is one of the things which is making it hostile to control, and perhaps that is in some part due to the original goals of the network.”

计算机网络技术的 历史和新进展