Network Fundamentals

Abstract

This chapter introduces the concept of network and Internet, common network topologies, network types, common networking equipment (routers, switches, firewalls and wireless devices), the design of Layer 2 enterprise LAN and Layer 3 enterprise LAN, and the concept of WAN.

This chapter introduces the concept of network and Internet, common network topologies, network types, common networking equipment (routers, switches, firewalls and wireless devices), the design of Layer 2 enterprise LAN and Layer 3 enterprise LAN, and the concept of WAN.

1.1 Overview of Network

1.1.1 Network and Internet

If you want to interconnect two computers, you can connect them directly through a network cable; however, if you want to interconnect three, four or even more computers, then network devices will be needed. As shown in Fig. 1.1, a network is be formed by connecting computers in close proximity to a switch via network cables.

Fig. 1.1

Network illustration

For the communication among multiple networks in different locations, they need to be connected through a router. Routers have WAN ports that can be used for long-distance data transmission and data packets forwarding between different networks. Multiple networks are interconnected by routers to form the Internet, as shown in Fig. 1.2.

Fig. 1.2

Illustration of internet

The Internet was originally formed by interconnecting the networks of universities and research institutions in the United States, and later more and more companies and government agencies were connected to the network. Gradually, the open network initiated in the United States was no longer limited within the country, and networks in a growing number of countries around the world are connected to this open network through submarine optical cables, satellites and so on, making it the largest global Internet at present, as shown in Fig. 1.3. Planning the network, configuring network devices and choosing the best path for data packets are the main and imperative tasks of network engineers.

Fig. 1.3

Illustration of the internet

1.1.2 Network Topology

Network devices (such as computers, routers, switches, etc.) are connected into different network topology through transmission media (such as twisted pair and optical fiber). Each network topology has its own advantages and disadvantages. According to the topological form, networks can be divided into star network, bus network, ring network, tree network, full mesh network, partial mesh network and hybrid network, as shown in Fig. 1.4.

• Star network: al nodes are connected to a central node.

  • Advantages: it is easy to add new nodes to the network. Communication data must be relayed through the central node, which makes it easy for network monitoring.

  • Disadvantage: if the central node fails, the communication of the whole network will be affected.

• Bus network: all nodes are connected to a bus (e.g., coaxial cable).

  • Advantages: it is easy to install and doesn’t require many cables; if a node fails, usually the communication of the whole network will not be affected.

  • Disadvantages: if the bus fails, the communication of the whole network will be affected; a message sent by one node can be received by all other nodes, which leads to low security.

• Ring network: all nodes are connected into a closed ring.

  • Advantages: it doesn’t require many cables.

  • Disadvantage: it is troublesome to add new nodes, which requires the original “ring” to be interrupted before new nodes can be added to form a “new ring”.

• Tree network: the tree structure is actually a layered star structure.

  • Advantages: it can quickly connect multiple star networks together, and can be layered as needed, which makes it easy for network expansion.

  • Disadvantages: the higher level the nodes are in, the more serious the network problems will be.

• Full mesh network: all nodes are connected to each other through cables.

  • Advantages: high reliability and high communication efficiency.

  • Disadvantages: each node requires a large number of physical ports, as well as a great many interconnecting cables, which is costly and difficult for network expansion.

• Partial mesh network: only the key nodes are interconnected with each other.

  • Advantages: lower cost than full mesh network.

  • Disadvantage: lower reliability than full mesh network.

• Hybrid network: this network topology is formed by combining the star, tree and partial mesh networks mentioned earlier.

  • Advantages: it has the characteristics of a star network, that is it is easy to increase nodes and monitor traffic in the center, as well as the layered characteristics of a tree network, and in the meantime, it is as reliable as a partial mesh network.

  • Disadvantages: redundant equipment and cables are required, which results in high cost.

Fig. 1.4

Network topologies

1.1.3 Network Types

Computer networks can be classified into different types according to the geographic coverage or users.

• 1. Classification by the network geographical coverage

  According to the network geographical coverage, computer networks can be categorized into Local Area Network (LAN), Wide Area Network (WAN) and Metropolitan Area Network (MAN), which is between a LAN and WAN.

  LANs usually cover an area of a few square kilometers, and its main function is to connect several terminal computers in close proximity (e.g., within a household, within a building or several buildings, and within a campus.). Generally, organizations purchase their own equipment to set up LANs. The technologies used in LAN include Ethernet, Wi-Fi, etc.

  WANs usually cover a geographical area of tens to thousands of kilometers, and are able to connect multiple cities or countries, or span several continents and meanwhile provide long-distance communication, thus forming an international remote network. For example, an enterprise has two LANs in Beijing and Shanghai, and when these two LANs are connected, a WAN is created. Generally, WAN requires to lease the line of Internet Service Provider (ISP) and pay a certain fee to ISP for bandwidth every year. The bandwidth is related to the fee paid. In the early days, households used Asymmetric Digital Subscriber Line (ADSL) to dial up to access the Internet, so there were different bandwidth standards such as 2 Mbit/s bandwidth, 4 Mbit/s bandwidth and 8 Mbit/s bandwidth, and corresponding charges. The technologies used in WAN include PPP, PPPoE and HDLC. The larger ISPs in China are China Telecom, China Mobile and China Unicom.

  A MAN is a larger LAN, which requires higher cost but can provide faster transmission. It improves the transmission media in LAN and expands its coverage to include a university campus or a city. Its main role is to connect hosts, databases and LANs in different locations within the same city. MANs are similar to WANs in terms of the role they play, but they differ in implementation and performance. A MAN is a large-scale LAN, using technologies such as Ethernet (10 Gbit/s, 100 Gbit/s) and World Interoperability for Microwave Access (WiMAX), which is similar to LAN technology.

• 2. Classification by users of the network

  According to their users, computer networks can be classified into public networks and private networks.

  A public network refers to a large network funded by a telecom company (state-owned or private). “Public” means that the network is available to anyone who is willing to pay for it as stipulated by the telecom company. The Internet is the largest public network in the world.

  A private network is a network built by a department for the special business needs of the organization. This kind of network does not provide services to people outside the organization. For example, military, railway, and electric power systems all have their own private networks.

  Both public and private networks are able to handle a variety of services. If they transmit computer data, they are called public computer networks and private computer networks, respectively.

1.2 Networking Equipment

Figure 1.5 shows a typical enterprise computer network, which looks a little complex, but can be seen as a network with a Layer 3 structure, i.e., access layer, aggregation layer and core layer. The only difference is that in order to avoid single point of failure, a high availability architecture with dual aggregation and dual core layers is applied. In addition, firewalls are deployed on the links to the Internet of the egress area, and the Internet is connected via dual links.

Fig. 1.5

Enterprise computer network

The devices in this network are switches, routers, firewalls, wireless devices, etc. The functions of various network devices are introduced in detail below.

• 1. Switch

  As shown in Fig. 1.6, in a campus network, the switch is generally the closest device to the end user. The network formed by an Ethernet switch is a broadcast domain, that is the broadcast frames sent by one node can be received by the rest of the nodes.

• 2. Router

  As shown in Fig. 1.7, a router is responsible for forwarding messages in different network segments, selecting a suitable path to deliver the message to the next router or destination based on the destination IP address of the message received, and the last router in the path is responsible for delivering the message to the destination host. Routers isolate broadcast domains, run routing protocols, build routing tables, maintain routing tables, forward IP messages, connect to WANs, perform network address translation and are connected to networks formed by switches.

• 3. Firewall

  A firewall is a network security device, as shown in Fig. 1.8, which is used to control secure communication between two networks with varying trust levels (e.g., the internal network of the enterprise and the Internet). It monitors, restricts and modifies the data flow crossing the firewall by formulating and implementing a unified security policy, so as to prevent external users of the network from illegally accessing important information resources inside the network. In other words, it shields the information, structure and operation status inside the network from those outside the network as much as possible, so as to provide security protection for the internal network of the enterprise. The main functions of a firewall are as follows.

  • Isolation of networks with different security levels.

  • Implementation of access control (security policies) between networks of varying security levels.

  • User identity authentication.

  • Implementation of remote access.

  • Implementation of data encryption and virtual private network services.

  • Performing network address translation.

  • Other security functions.

• 4. Wireless devices

  Common wireless devices are access points (AP) and access controllers (AC).

  • Access points

    An access point is used for a wireless switch of the wireless network, and is also the core of the wireless network. An access point enables mobile computer users to access the wired network, and is mainly used for broadband home, inside the building and campus. It typically covers a distance ranging from tens of meters to hundreds of meters.

    Access points generally support three working modes, namely, fat AP, fit AP and cloud AP, as shown in Fig. 1.9. According to the demand of network planning, various modes can be chosen flexibly.

    Fat AP: it is suitable for home, as it works independently, requires separate configuration, provides single function, and is low-cost.

    Fit AP: it is suitable for large and medium-sized enterprises, as it needs to be used in conjunction with ACs, requires the unified management and configuration by AC, and provides rich functions.

    Cloud AP: Cloud AP is interconnected with cloud management platform and covers tens of thousands of square meters. With the help of cloud management platform, cloud AP is capable of plug-and-play deployment and cloud-based remote operation and maintenance in small and medium-sized simple networks and large complex networks. The “cloud management platform + Cloud AP” networking model has become the preferred option for enterprises with a multitude of branches, such as retail stores, small and medium enterprises as well as hotels.

  • Access controllers

    An access controller (AC) is an access point control device of wireless LAN, which is responsible for converging and connecting data from different APs to the Internet, and performing the configuration management of AP devices and control functions such as authentication, management, broadband access and security of wireless users.

    Access controllers are generally located in the aggregation layer of the whole network, providing wireless data control services with high capacity, high performance, high reliability, easy installation and easy maintenance. It has multiple advantages, such as flexible networking, green and energy saving.

Fig. 1.6

Switch

Fig. 1.7

Router

Fig. 1.8

Firewall

Fig. 1.9

Access points

1.3 Planning and Design of Enterprise Networks

Enterprise networking involves LANs and WANs. The deployment of network devices for the enterprise LAN should be combined with the physical location of computers. Generally, switches are used for the connection within the LAN while routers for the connection of WAN.

1.3.1 Enterprise LANs

The concepts of network and Internet are introduced earlier, but networks that most people are familiar with are those at their homes and in enterprises. Depending on the network size and the physical distribution location of computers, enterprise networks can be designed with a Layer 2 or Layer 3 structure, and are usually designed with a tree or a hybrid network topology. A Layer 2 network has limited networking capability and is usually only a small LAN; and a Layer 3 network can form large networks. The following are two typical scenarios to demonstrate the design of Layer 2 and Layer 3 enterprise networks.

• 1. Layer 2 LAN

  Here is an enterprise network as an example to introduce the network topology of an enterprise LAN. As shown in Fig. 1.10, first, a switch is deployed in Office 1, Office 2 and Office 3 to connect the computers in the offices. The switches in the offices require many ports so that more computers can be connected to the network. The switches at this level are called access layer switches, which are, at present, typically connected to computer ports with bandwidth of 100 Mbit/s.

  The aggregation layer can be deployed with either a router or a switch. If a switch is deployed there, it is usually a Layer 3 deployment that performs IP message forwarding tasks.

  A switch can be deployed in an enterprise server room to connect servers in the enterprise to switches in each office, so as to converge Internet flow from access layer switches in the offices, and Internet connection is enabled through a router. Switches at this level are called aggregation layer switches. The switch at this level does not need many ports, but the port bandwidth should be higher than that of the access layer switch, otherwise it will be a bottleneck that limits the network speed.

• 2. Layer 3 LAN

  In enterprises with larger network size, a Layer 3 structure may be adopted for LAN. There are three levels of switches in a Layer 3 LAN: access layer switches, aggregation layer switches and core layer switches. The layered model can be used to help design layered networks that are expandable, reliable, and cost-effective.

  As shown in Fig. 1.11, an enterprise has three branches, each of which has its own office building with its own server room and network. The network center of this enterprise provides Internet access for the three branches, so the aggregation layer switches of each branch have to be connected to the switches in the network center, and the switches at this level are called core layer switches. The enterprise’s servers are connected to the core layer switch to provide services to the three branches.

  The aggregation and core layers can be deployed with routers or switches. If a switch is deployed, it is usually a Layer 3 deployment that performs IP message forwarding tasks.

Fig. 1.10

Layer-2 LAN

Fig. 1.11

Layer 3 LAN

1.3.2 Enterprise WANs

In the networking process, in addition to LAN, enterprises may also use WAN. As depicted in Fig. 1.12, the vehicle factory has plants in both Shijiazhuang and Tangshan, and the vehicle factory of CSR in Shijiazhuang and that of CNR in Tangshan have formed their own networks. It can be seen that the vehicle factory has planned the network according to the department (such as vehicle assembling workshop, wood company, transportation company, mechanical and electrical plant in the figure). Basically, each department has a network (network segment), and a core layer switch (equivalent to router) is used to connect the network segment of each department, that is, the enterprise’s server is connected to the Layer 3 switch. This is the enterprise’s LAN.

Fig. 1.12

LAN and WAN

At this time, if the CNR vehicle factory in Tangshan needs to access the server of CSR vehicle factory in Shijiazhuang, the networks of the two factories need to be connected. It is unlikely for the vehicle factory to connect the LANs of these two factories by setting up network cables or optical fibers by itself, as this will lead to a sky-high cost for construction and maintenance. Enterprises usually connect the two LANs by leasing the carrier’s line, so that they only need to pay the annual fee, thus forming the enterprise WAN.

In general, an enterprise LAN is usually formed by the enterprise by paying for network devices by themselves with a usual bandwidth of 10 Mbit/s, 100 Mbit/s or 1000 Mbit/s, which is maintained by enterprises themselves and covers a small area. An enterprise WAN is usually set up by enterprises by paying for the leased lines of carriers, i.e., they pay for the bandwidth so as to achieve long-distance communication.

1.4 Exercises

• 1. What is a network? What is the Internet? What is the largest interconnected network at present?

• 2. List some common network topologies and explain what kind of network topology is usually used for enterprise networking.

• 3. List some common network devices and state their roles.