计算机网络I-考前指导复习提纲

计算机网络I考前指导

填空、选择、连线、简答、计算等题型

考试范围基本在course review里面，老师说可能会考的用表示（自己学校的考点都在这里，仅供参考）

PPT1

Networking basics

1. What are the differences between data networks and voice networks（传统的）?

   要求：要区分哪些特点是属于data networks/voice networks，判断/选择

   1. data networks:(1)communication：use digital transmission(2)data：encapsulated into datagram（3）send : through network devices（4） discrete signal（离散信号）. eg.Ethernet network
   2. voice networks:(1) communication：uses analog transmission(2)voice stream: are forward by circuit（4） continous signal. eg.telephone network

2. How many types of computer networks when it is classified by their scale?

   4 kinds：Personal Area Networks, Local Area Networks, Metropolitan Area Networks, and Wide Area Networks.个人区域网、局域网（公司、校园eg.以太网）、城域网和广域网(城市、国家)

Network architecture

1. What are the definitions of layers, protocols, and services in network architecture?

   ▪ A layer is a level of abstraction which has a well defined function.（抽象概念，每一层定义了功能，同一层功能一样）

   ▪ Protocols are executives([ɪɡˈzekjətɪvz]) of the functions.（实现功能的载体、执行，协议是一组指定设备如何交互和交换消息的规则）

   ▪ Services are functions provided through protocols.（协议产生的功能）

2. What are the famous reference models used in computer networks?

   1. OSI
   2. TCP/IP

3. What are the differences between OSI and TCP/IP reference model?

   以下为比较：

   ▪ Both have layers. But TCP/IP appears simpler because it has fewer layers.

   ▪ Both have application layers, though they include very different services. TCP/IP combines the

   presentation and session layer issues into its application layer.

   ▪ Both have comparable transport and network layers. But TCP/IP combines the OSI data link and

   physical layers into the network access layer.

   ▪ Both assume packets are switched. This means that individual packets may take different paths to

   reach the same destination.

   ▪ The TCP/IP model gains credibility just because of its protocols. In contrast, the OSI model is used

   as a guide.

   ▪ 两者都有层次。但是TCP/IP看起来更简单，因为它的层数更少。

   ▪ 两者都有应用层，尽管它们包含非常不同的服务。TCP/IP将表示层和会话层问题结合到应用层。

   ▪ 两者都有相当的传输层和网络层。但是TCP/IP结合了现场视察数据链路和

   物理层进入网络接入层。

   ▪ 两者都假设数据包被交换。这意味着单个数据包可能通过不同的路径到达同一个目的地。正是因为它的协议，TCP/IP模型获得了可信度（被用了）。相比之下，OSI模型被用作指南（没被用）。

Functions and technologies in OSI and TCP/IP reference model

1. What are the names of seven layers in OSI and TCP/IP reference model?
   
2. What functions are defined in each layer?

要求：OSI、TCP每一层的功能（OSI）、协议（TCP/IP）、哪些地址（TCP/IP）定义了

考法：某一层是什么功能

1. OSI reference model:

   (1)Application layer:Data exchange between network applications.

   (2)Presentation layer:Data representation between hosts.

   (3)Session layer:Dialogue control between hosts.

   (4)Transport layer:Reliable transmission between hosts.

   (5)Network layer:Best path determination between hosts.

   (6)Data Link layer:Media access and reliable transmission over physical

   link.

   (7)Physical layer:Bits transmission over physical link.

   网络应用程序之间的数据交换。
   主机之间的数据表示。
   主机之间的对话控制。
   主机之间的可靠传输。
   主机之间的最佳路径确定。（寻址、路由）
   物理链路上的媒体访问和可靠传输。
   物理链路上的比特传输。

2. TCP/IP:

▪ The application layer should include the OSI application, presentation and session layer details.

They created an application layer that handles issues of representation, dialog control, and etc.

▪ The transport layer deals with the issues in the quality of transportation services, such as reliability, flow control, and error correction.

▪ The internet layer is to divide segments into packets and transmit them between different networks.

Best path determination and packet switching occur at this layer.

▪ The network access layer, also known as the host-to-network layer, is concerned with all of the

components that are required to make a physical link. It includes the details in the OSI physical and

data link layers.

▪ 应用层应包括OSI中的应用、演示和会话层细节。他们创建了一个处理表示、对话控制等问题的应用层。

▪ 传输层处理运输服务的质量问题，如可靠性、流量控制和纠错。

▪ 互联网层是将数据段分成数据包，并在不同的网络之间传输。最佳路径确定和分组交换发生在这一层。

▪ 网络接入层，也称为主机到网络层，负责建立物理链路所需的所有组件。它包括OSI物理层和数据链路层的细节。

3. What protocols are defined in each layer?

   考法：eg.IP协议是在TCP/IP哪一层定义的、Domain Name（某一个地址）是在TCP/IP哪一层定义的？

   主要关注TCP/IP：

   （1）The application layer protocols include:

   • Domain Name System (DNS)

   • Hypertext Transfer Protocol (HTTP)

   • File Transfer Protocol (FTP)

   • Simple Mail Transfer Protocol (SMTP)

   • Trivial File Transfer Protocol (TFTP)

   （2）The transport layer protocols include:

   • Transport Control Protocol (TCP)

   • User Datagram Protocol (UDP)

   （3）The Internet layer protocols include:

   • Internet Protocol (IP)

   （4）The network access layer refers to any particular technology used on a specific network (Ethernet).

   (1)应用层协议包括:

   域名系统

   超文本传输协议

   文件传输协议

   简单邮件传输协议

   普通文件传输协议(TFTP)

   (2)传输层协议包括:

   传输控制协议

   用户数据报协议

   (3)互联网层协议包括:

   互联网协议

   这些都要知道，都是底层往上层服务

   (4)网络接入层是指在特定网络(以太网)上使用的任何特定技术。
   
   图中值得注意的是PDU（协议数据单元）头：协议，有效载荷：上层PDU，尾：数据校验和

   分别是Data/segment/packet/Frame,Bit数据/段/包/帧，位

4. What addresses are defined in each layer?

   每一层的地址：

   domain name域名：应用层(application)

   port端口第四层（OSI的）(transport)

   IP地址是第三层(network)

   MAC地址：第二层(TCP/IP的network access)

5. What network devices are operating at each layer?

   hubs,switch,router,application

6. The technologies of LANs and WANs are usually designed in which layers?

   这些LANs 和 WANs是技术，用1、2层定义的。

Data encapsulation

1. What is the process of data encapsulation?

   知道了这么多设备、每一层的功能，可是它是怎么用这个模型传输信息的，怎么将功能组成起来的，就是用数据封装。

   要求：了解封装是数据包越来越大的过程，解封装是数据包越来越小的过程；发送是封装，接收时解封装；发送端从应用层往下走，接收端则往应用层往上走；封装和解封装一共经历多少层：由设备所在层数决定；判断两台设备要通讯哪一层和哪一层协议一样才能通讯（不同层协议能通讯有什么要求）：如果一个设备和另一个紧挨着的设备有第一层协议，则这两个的第一层协议要一样(相邻的设备的同一层协议要一样）。

   （1）封装：从上层到底层封装，数据被包裹在PDU中，往下每一层都加上下一层的Header和Trailer，所以逐层PDU会越来越大。

   （2）解封装：底层到上层，把每层的Header和Trailer去掉给上层。

   由此，每一层的功能都会被用到，并且达到每一层只会看到对应那一层，即同一层的协议只会跟同一层通讯，两台设备要相同的协议才能通讯（解封装、解到某一层时候把那一层协议交给对应的设备同一层协议去处理）。

   注：Encapsulation and de-encapsulation between networks focuses ONLY on layers 1, 2, and 3 of the OSI model.

2. What PDUs are used in each layers?

   考法：TCP属于哪一层的PDU？TCP是OSI的第四层，它的PDU是segments。这里PDU是统称，就比如有IPV４的packet也有IPv６的packet。

   (1) Application layer:data.

   (2) Transport layer:segments.

   (3) Internet layer:packets.

   (4) Network Access layer:frames and bits.

3. What are the header format of common used PDUs in each layer?（这个了解就好，不用背了）

Network topology and terminologies

1. What are the differences of physical and logical network topologies? List some common used physical and logical network topologies.（了解就好）

   ▪ Network Topology defines the structure of the network. There are two parts to the topology　definition:

   ▪ Physical topology, which is the actual layout of the wire (media).

   ▪ Logical topology, which defines how the media is accessed by the hosts.

   网络拓扑定义了网络的结构。拓扑定义分为两部分:

   物理拓扑，即导线(介质)的实际布局。

   逻辑拓扑，定义主机如何访问介质。

   网络可以有一种物理拓扑，也可以有完全不同的逻辑拓扑。

2. What is the unit of digital bandwidth and analog bandwidth?（了解单位）

   前者，digital bandwidth逻辑的带宽：ｂｐｓ，后者，analog bandwidth物理的带宽：Ｈｚ（震动多少次）

   bps:bps是什么单位？是什么意思_NCTU_to_prove_safety的博客-优快云博客_bps是什么意思

3. What is the definition of bandwidth and throughput, and baud? What is the relationship among

   bandwidth, throughput, and baud?

   (a)bandwidth:maximum theoretical amount of information that can flow through a network connection in a given period of time.

   (b)throughput:actual measured bandwidth, at a specific time of day, which is affected by factors that include number of users on network, networking devices, type of data, user’s computer and the server, etc.

   ©baud:The time required to transmit the character depends on both the encoding method and the signaling speed. The number of changes per second is measured in baud.

   (a)带宽：在给定时间内可以通过网络连接流动的最大理论信息量。(b)吞吐量：在一天中特定时间的实际测量带宽，受网络上的用户数量、网络设备、数据类型、用户计算机和服务器等因素的影响。©波特：传输字符所需的时间取决于编码方法和信令速度。每秒的变化次数以波特为单位测量。

   (d)关系：bandwith是理想的speed，throughput是实际的speed，baud是另一种方法测量bandwith.

   Bandwidth is the theoretical maximum information and throughput is the actual measured bandwidth.Throughput ≤ Bandwidth.

   考法：考运算eg.假设一个符号有8种变化（L），每秒钟发送3个波特/每秒钟（发送3个信号波特率为3)。但是现在有4种变化，问实际的逻辑带宽（digital bandwidth）有多少bps？bit rate = baud * log2L bit rate=3*2=6bps 注:考试不一定是整数，若不是，则直接写出公式eg.bit rate=3 *log2 10

   波特（率,即Baud）和比特率关系：波特率与比特率_一剑钦安-优快云博客_比特率和波特率，文中的符号就是这里的信号。一个波特=一个信号=按照具体编码的几位。比特率：是单位时间内传输了多少位（比特），单位是bit/s或者bps（bits per second比特/每秒)。比特率 = 波特率 * （一个符号）几位。如果一个信号有2位，则有4种变化（00 01 10 11），一个符号有3位就有8种变化。

   注：L的另一种说法：这里的L是用于传输的电压等级Here, the L is the voltage levels used in transmission.

4. What is the difference between the transmission delay and propagation delay?

   考法：定义和计算都要考。

   (1)Transmission delay: is the time needed to push all the packet bits on the transmission link. It has nothing to do with the distance between the routers,but related to packet length and link transmission rate.

   (2)Propagation delay: is a simple measurement of how long it takes for a signal to travel along the cable being tested.And it is the time it takes for a signal to travel from one router to another in the propagation medium.So it is related to the distance between routers.

   （1）传输延迟：是推送传输链路上所有packet bits所需的时间。它与路由器之间的距离无关，而是与数据包长度和链路传输速率有关。信号传播的速度

   （2）传播延迟：是一个简单的测量信号沿着被测试的电缆移动所需的时间。它是一个信号在传播介质中从一个路由器传输到另一个路由器所需的时间。所以它与路由器之间的距离有关。cable的长度

   计算：

   eg.考经过几个设备以后总的时间是多少：

   The sender and receiver are separated by two 1 Gbps links and a single switch.
   The message size is 5000 bits, and each link introduces a propagation delay of 10 us. What is the total transfer delay?
   (Let’s assume that the switch begins forwarding immediately after it has received the last bit of the message.)

   发送方和接收方由两个1Gbps链路和一个交换机分开。消息大小为5000位，每个链路引入了10us的传播延迟。总传输延迟是多少？（让我们假设交换机在收到消息的最后一点后立即开始转发。）

   注：这里的10⁹是要注意一个单位的换算，TD和link transmission rate有关，所以给的链路速度是用