1.1 Define common networking terms for LANs and WANs.
 

According to Microsoft's definition, The Networking Essentials certification exam measures your ability to implement, administer, and troubleshoot information systems that incorporate Microsoft® Windows® 95 and any products in the Microsoft BackOffice family. Please note there are two important points in the above definition, one is related to networking concepts, and the other is related to Microsoft Window systems. In fact, neither point can be ignored for your preparation. Some people complain that Microsoft's Networking Essentials exam is a tough one. They could not pass the exam even they had many years experience in networking. In fact, they ignore the knowledge on Windows products, which is also very important for the exam.

A successful strategy for preparation would be to study the networking items that are Microsoft related. It will definitely save you a lot of time. For example, There are more than five RAID implementations: RAID 0, 1, 2, 3, 4, 5, 7. Study all of them will cost you lost of time, but if you know Windows NT server only supports RAID 0,1 and 5, you can just study these three RAID technologies.

Microsoft will NEVER test you the technologies they are not supporting. Otherwise, they are stupid. For example, if someone had learnt a lot on RAID, they might try to implement RAID 4 on NT. However Microsoft cannot provide the solution for them, and they have to contact the hardware manufacture whose equipment supports RAID. Isn't it ridiculous?

Microsoft suggests Networking Essentials as the first exam for the MCSE exams because almost all the MCSE exams need the networking background. Some people dislike Networking Essentials exam because it is somewhat tedious compared to other exams such as Windows 95 and Windows NT. On those exams, you can practice while study. However, the Networking Essentials exam has all the concepts. Therefore, my tutorial will let you only focus on the concepts that will be tested. It is always correct '
What will be tested, what should be studied'.

OK, let's begin with some basic networking concepts: network, NIC, LAN and WAN. Please note Microsoft will not test you these concepts directly. Microsoft will NEVER let you provide the definition of LAN and WAN on the Networking Essentials exam. However, these concepts may appear in the exam as well as your real work. If you have no idea on these concepts, it will prevent you from understanding the whole question.

 

Network: Two or more computers that communicate with each other. The goal of the network is sharing information between computers. Figure 1.1 shows a simplest network, which has only two computers.

 

  Figure 1.1 the simplest network

Computer, NIC and network cable are the three elements of a network. A more complicated network may involve more network devices such as Hubs and repeaters etc. However, computers, NICs and network cables are necessary to build a network.

NIC: Network Interface Card, sometimes refers to as Network Adapter. NIC (Figure 1.2) is plugged into the slot of the computer and provides the physical connection between the network cable and the computer.

Suppose NIC A is plugged into computer A, while NIC B is plugged into computer B. If a user of computer A wants to send a message to a user of computer B. The message goes to the NIC A first. After that, NIC A converts the message to small packets that can be transmitted via the network cable. After NIC B receives the packets from NIC A, it reassembles them and converts the data back to a format that can be understood by the computer has NIC B installed. This mechanism makes the communication possible between two computers.

Figure 1.2 NIC

LAN: Local Area Network. LAN is a network that covers a relatively limited area, such as an office or a building.

WAN: Wide Area Network, is a network that connects geographically separated areas.

Sometimes it is difficult to distinguish a LAN from a WAN. If somebody ask you following question, 'Is connecting two networks in two buildings a LAN or a WAN solution?' You probably cannot answer this question immediately. If these two buildings are only 100 meters apart, it can be a LAN. While if one building in Syracuse and one building in New York city, it must be considered a WAN solution.

In fact, what distinguishes a WAN from a LAN is not important for us. We care more about whether a technology can be used in a WAN or a LAN environment. For example, Ethernet represents a typical LAN solution while telephone network is a typical WAN solution. Only ATM is the only technology that can be deployed in both LAN and WAN.

If you didn't have any experience on the networking before, you might have some difficulties on understanding those networking concepts I mentioned above. But don't worry, let's begin with some real world networks and hope you can have more confidence on the preparation.

Ethernet should be the most popular LAN network access method in the world. Xerox Corporation introduced the first Ethernet product in 1975. At that time, the Ethernet only provided 2.94Mbps-transmission speed. Ethernet was so successful that Xerox, Intel® Corporation, and Digital Equipment Corporation (now Compaq) drew up a standard for a 10-Mbps Ethernet. The first 10-Mbps Ethernet product was shipped in 1981. In 1983, the IEEE approved the Ethernet 802.3 standard that we'll discuss later.

We are usually using bps to measure the network transmission speed. bps means bits per second. 1 byte = 8 bit. We normally use capitalized B stands for bytes and b for bits. Therefore, 1Bps equals 8bps for the communication. In fact we use byte while not bit for the measurement in many computer areas. For example, if you open the Windows Explorer on a Windows 95 computer, you will find the file size information is displayed in byte basis while not the bit. I think the only reason why networking people prefer to use bit in that measuring in bits gives you larger number and lets people think the network is fast. For example, 10Mbps Ethernet only equals to 1.25MBps. If people saw 1.25MBps Ethernet, they might think the Ethernet is slow. Understanding bps can also help you understand how fast the transmission speed. Suppose you have a 10Mbps Ethernet, it is very quick to copy files over the network. Suppose now you work at home and only have 56Kbps-modem connection for the Internet. If you need to download a 560KB (560KB=560*8kb) word document over the Internet, you at least need to wait 560 * 8 /56 = 80 seconds. This is just a rough estimation. In most case, to transmit 1 byte of data requires 10 bits (a start bit and a stop bit are needed for each 8 bits of data). At the higher level of aggregations, there are headers and protocol flags to deal with. Therefore, to download 560KB of data over the modem from the Internet, you need to wait at least 560000 * 10/56000 = 100 seconds.

In the real world, if you saw a network follows the Ethernet standards, you can call it an Ethernet. For the networking essential exam, you must know that Ethernet uses CSMA/CD technology to access the network.

CSMA/CD, acronym for Carrier-Sense Multiple Access with Collision Detection. Carrier Sense means that the network card listens to the cable for a quiet period during which it can send messages. Multiple Access refers to multiple computers using the same network cable. Collision Detection is a safeguard against messages colliding in transit. If a computer want to send a packet over the network, it listens to the network to see if there are any other packets are transmitting. A packet can be only transmitted when no other traffic on the network. But it still might happen that two listeners find the network is idle and send the packets together. At this time, you need to use CD (Collision detection) to prevent the conflict.


Ethernet normally shares the bandwidth, which is either 10 Mbps (Ethernet), 100 Mbps (Fast Ethernet) or 1000 Mbps (Gigabit Ethernet). With switched Ethernet, each sender and receiver pair has the full bandwidth. Therefore, switched Ethernet improve a lot on the performance and more and more Ethernet networks moved to switched Ethernet. Some people might still have some difficulties on understanding how a switched Ethernet works. They can understand Ethernet shares the bandwidth because it has collision detection to prevent multiple stations send the packet together. All computers in the network share the same media as well as the bandwidth. But how come switched Ethernet each receiver and sender can establish the connection with full bandwidth? In fact, the backbone bandwidth on an Ethernet switch is very high. For example, 10Mbps Ethernet switch might have 100Mbps backbone bandwidth, which guarantees the 10Mbps connection between the receiver and the sender.

Token Ring is the next most popular LAN access method than Ethernet. In a token ring network, a token is passed around the ring to which the computers are connected. The computer that grabs the token is able to send data over the network. As we discussed before, in an Ethernet, all computers try to send the network packets, collision detection is required. However for the Token Ring, only the computer holds the token can send the packets, which is quite simple and efficient. After Microsoft changed the exam to adaptive, they consider Token Ring questions as tough questions. By my estimation, Microsoft will keep adding Token Ring questions in the MCSE exams. So be careful on your preparation.

Before we study more materials, let's give you some tips on the preparation. I think the most important one is taking your time on your preparation. Some people read the questions and the answers from beginning to end. They just try to remember the answer for each question. It is NO good.

We have found that Microsoft sometime only modifies something on the question and the meaning of that question is totally changed. At this time, if you can not understand why the answer is on FTQ_7058, you may miss the point on the exam. In your preparation, Accuracy is ALWAYS more important than speed. I ALWAYS ask people to try understanding why the answer is when they are reading FTQ_7058.

For the traditional fix-length exam, you need to give answers as quickly as possible and do not get hang-up on one difficult question. However Microsoft changed this exam to adaptive now. Please note Microsoft still use adaptive technology in its new fix-length exam. You need to make sure that your answer is correct before you press the 'Next' Button. You don't have any chance to correct the answers anymore.

In the preparation, you need to understand everything that appears on the question. As I mentioned on Microsoft Psychology, if you saw one term on the questions, it means Microsoft is testing this term. Even it is not the correct answer for this question, it may be the correct answer for other questions. Therefore, you should know any term that appears on the question. For example, although only answer C (Cross talk) is the correct answer for FTQ 1.1.1, you should not skip studying other terms such as Attenuation, Loop back as well as Jitter.

Attenuation means signal decreasing in volume over a distance. That's the reason why networking standard specifies the distance limitation for networking cables. For example, ThinNet coaxing cable has the distance limitation of 185 meters. If you use a ThinNet coaxing cable connects two computers over 185 meters, the signal cannot be guaranteed to reach the other end because of the attenuation feature of the cable.

Loop Back redirects the outgoing signals back into the computer for testing. Figure 1.3 shows a 25-pin serial port and a loopback plug.

Figure 1.3 A 25-pin serial port and a loopback plug

Before we explain more on the concept of loop back, let me ask you one question: 'What's the most efficient way to copy one big file (Say around 30MB) from one computer to another?'

Some people may consider using 1.44M floppy disk. General copy command doesn't work now. You need to use backup and restore command along with several floppy disks. If these two machines are nearby, one floppy disk is enough, you can reuse this disk for backing up and restoring. If two machines are far way, you need 30/1.44 = 21 floppy disks to copy a file 30MB in size.

Some people may think of using Zip disk driver from Iomega. A Zip driver works like a floppy disk and has 100MB capacity. It may replace floppy disk in the future. However now, it is optional and you need extra $99 to purchase a Zip disk driver.

The solution using network card can also be considered. If your system has a network card pre-installed. It should be the best solution. If not, you still need extra money to purchase network card and network cables.

In fact, the most efficient way is to use a serial cable. We know almost all PCs have one to two serial ports, you just need a cable and corresponding software to make the connection and copy files from these two machines. It is also called Direct Cable Connection in Microsoft's terminology. To install a Direct cable connection in Windows 95, follow these steps:

1. In the Add/Remove Programs icon in Control Panel, click the Windows Setup tab.
2. In the Components list, click Communications, and then click the Details button. 3. In the Communications dialog box, click Direct Cable Connection, and then click OK.
 

Windows 95 provides a Direct Cable Connection wizard for establishing the connection between two computers. The wizard runs when you open Direct Cable Connection for the first time. It allows you to designate one computer as the guest and the other as a host. Before you run the wizard, you need to install Direct Cable Connection on each computer and connect them with a null-modem serial or parallel cable.

Some pins of the serial port are used for receiving data while other pins are for sending data. If you transfer data successfully from one computer to another with serial cable, it means the serial port at each end works properly.

However, if you only have one machine, how do you know your serial port has no problem? At this time, a loopback plug shown in figure 1.3 can be used. A loopback plug can be plugged into a port, such as a serial or parallel port, and crosses over the transmit line to the receive line so that outgoing signals can be redirected back into the computer for testing.

Sometimes you may see loopback address in TCP/IP network. Here loopback address means IP address 127.0.0.1. After you install the network card driver and TCP/IP protocol, you can use 'ping 127.0.0.1 ' command to test if the software is installed successfully. This command routes outgoing packets back to source computer for testing.

Crosstalk means interference from another physically adjacent medium. It may cause the data corruption. If you put a network cable and a power cable as in figure 1.4, crosstalk may happen.

Figure 1.4 Crosstalk

Crosstalk may happen for most of the networking cables when two cables are close to each other. Only the optic fiber is crosstalk free. Therefore, in some noisy environment, fiber optic is the only choice. If you are designing a network for your customer, you also need to be aware crosstalk happens. That's the one reason why people put power cables on the roof while network cables under the floor. That's not only for safety but also to avoid the crosstalk.

Beaconing means continuous signaling of error conditions on a LAN. However, in Microsoft's networking essential exam. Beaconing has a different meaning. Before we start to learn the definition of beaconing, let you give you some tips in the exam. First of all, when you see both Token Ring and Beaconing on the exam questions and answer choices, you should choose beaconing as the correct answer. That's because beaconing is Token Ring related. Microsoft doesn't test much token ring in the networking essential exam. Therefore, if you see both token ring and beaconing appear in the question, just choose beaconing. If you want to know more on beaconing, here is the definition:

Beaconing is defined as the process that is used to determine the networking states and fault location in case of a cable fault on a Token Ring network. One may also encounter beaconing when an advanced cable tester is used. This instrument can monitor the network and provide count, collision and beaconing information.

Jitter means a flickering in transmission signal. Both Jitter and crosstalk cause the errors in the accuracy of the signal. Generally, jitter only last a few seconds while crosstalk can be happen on a continuous basis if the cable is in crosstalk status.By the way, jitter only happens on broadband network, which we will discuss later.

Latency refers to the time between initiating a request for data and the beginning of the actual data transfer. For example, if you want to send a packet, it cannot be sent immediately as some time is needed for initialization. The time interval required for initialization is called latency time. On disk operation, latency is the time to locate the sector position.

Both jitter and crosstalk produce error signals while latency only add delay to the transmission.

Jam Packet is the packet that jammed in communication. Generally it is because the buffer size is not adequate. Jam packet may cause the whole network to crash.

Signal-to-noise ratio measures the clarity or quality of a transmission channel, audio signal or electronic device.

Signal-to-noise ratio = amplitude of a data signal/amount of noise

Broadband & Baseband To understand broadband and baseband, one needs to know two basic terms first: TDM and FDM.

TDM stands for Time Division Multiplexing. You can think of TDM as Time Division plus Multiplexing. Multiplexing means that you can transmit multiple signals simultaneously over a single transmission path. Time Division means that you slice the time in order to transmit different signals. TDM technology is used in T carries such as T1, T2, which is very popular in WAN environment.

FDM means Frequency Division Multiplexing. Same as TDM, FDM also transmits signals simultaneously over a single transmission path, such as a cable or wireless system. However, TDM slice the time while FDM separate signals by its own unique frequency range. Typical example of FDM is cable TV system. Although you only have one cable (means one channel), you can receive several TV channels. Each channel has a different frequency range. That is the reason why when you switch to channel 3, you go to NBC, and to channel 5 you go to CNN…

Some people may still have some confusion on FDM. They think each TV channel signals should be similar and you cannot distinguish them by frequency. It is true that there isn't have much difference from one TV signals to another. For example, the frequency may always be in the 1000Hz to 1800Hz range. If you put all these signals without conversion on the transmission cable, the ender user's screen will become mess.

However, here comes a solution. For channel 1, we just transmit the raw signals. While for channel 2 we add 1000HZ on all the signals, therefore, the channel 2's frequency is in the 2000Hz to 2800Hz range. For channel 3, we add 2000Hz, and you will never get the channel overlap for the transmission. At the user end, if you choose station 2, we only receive the signals between 2000Hz ~ 3000Hz, we ignore all the other signals and reduce the signal by 1000Hz for the final output signals. After that, you can watch the TV with the channel you select. This kind of conversion is also called modulation in networking terms.

Suppose we have three kind of signals as shown in figure 1.5.

Figure 1.5 Three different signals

For the TDM, we slice the time into small intervals as in figure 1.6.

Figure 1.6 TDM

After you divide the time to T1, T2, T3 etc, you can assign each period to transmit one signal. For example, T1, T4, T7 transmit signal 1, T2, T5, T8 transmit signal 2, T3, T6, T9 transmit signal 3, there will be no conflict on the transmission. Because the channel generally has much higher speed than the incoming signals, it can transmit the data without extra delay.

For the FDM, we convert the frequency of incoming signal to guarantee no conflict with them and transmit the all the signals at the same time.

Figure 1.7 FDM

Baseband A communications technique in which digital signals are placed onto the transmission line without change in modulation. If you study TDM and FDM carefully, you will find TDM doesn't need modulation because it just put the signal into a specific time interval. While FDM needs modulation because it put all the signals together for the transmission. Therefore, Baseband is using the TDM technology. For the networking essentials exam, you also need to know Token Ring and CSMA/CD use baseband technology.

Broadband A technique for transmitting data, voice and video using FDM. Modems are required for this method, because the digital data has to be modulated onto the line. The typical example of broadband is cable TV as well as radio station.

For the networking essential exam, there are three things you must remember:

1. Baseband refers to single frequency while broadband refers to multiple frequency ranges.
2. Baseband is bi-directional while broadband is unidirectional.
3. Baseband uses repeater to regenerate the signal while Broadband uses amplifier to regenerate signals.
 

Even if people know the definition of broadband and baseband, they may still mix them up because these two terms are too similar. Sometime people may choose broadband when Microsoft is talking about baseband. It is also time-consuming to figure out which one us during the exam.

To distinguish the characteristics of broadband and baseband, let me introduce another term broadcasting to you. Everybody knows that broadcasting related to TV station and radio. Now, try to remember the following statement: broadband supports broadcasting. Because broadband supports broadcasting such as cable TV, broadband needs to send multiple frequency at the same time (This allows you to select from multiple channels from your radio or TV). It is unidirectional, which means only TV, cable or radio station can delivery signals to you, while you can not send the information back.

After that, you can think of the baseband as the direct opposite of broadband. This way you will NEVER mix up these two terms in the exam.

FTQ 1.1.2 tests you the knowledge on network protocols.

What is network protocol? An analogy would help explain protocols: if you think of a PC as a person, network protocol is the language that person speaks. Only two people that speak the same language can communicate with each other. The same thing can be said about computers. Only two computers have the same network protocol can communicate each other. As a person can speak several languages, a computer can have several network protocols installed and running.

The three most important protocols for Networking Essentials exam are NetBEUI, TCP/IP and IPX/SPX.

NetBEUI (NetBIOS Extended User Interface) was introduced by IBM for small size LANs in 1985. At that time, Microsoft and IBM worked together on OS/2, which was planned as the replacement of the DOS system.

Although Microsoft switched to its Windows system and left IBM alone with OS/2, NetBEUI was kept for Microsoft's operating system such as Windows for Workgroup, Windows 95 as well as Windows NT. NetBEUI is very efficient for a small size LAN. The only problem is that NetBEUI doesn't support routing. Therefore, in large size LANs or the WANs such as the Internet, you must choose a routable protocol such as TCP/IP and IPX/SPX.

The fact that NetBEUI isn't routable doesn't mean that Windows NT doesn't support routable network, because you can also install TCP/IP on a Windows NT machine.

TCP/IP (Transmission Control Protocol/Internet Protocol) is a communications protocol developed for WANs. It was developed in 1969 under contract from the U.S. Department of Defense to internetwork dissimilar systems.

Originally, DARPA (Defense Advanced Research Projects Agency) developed TCP/IP to connect its research networks. This combination of networks continued to grow and now includes many government agencies, universities and corporations. This global WAN is called the Internet. Now TCP/IP is the protocol of the Internet and widely supported. TCP/IP is a routable protocol, which means it can route the packets from one LAN to another. This feature makes it possible to build VERY large network such as Internet.

IPX/SPX (Internetwork Packet Exchange/Sequenced Packet Exchange) are transport protocols used in Novell NetWare networks. Microsoft implements IPX through NWLink in its Windows Operating systems such as Windows 95 and Windows NT.

Some people may ask following question, "NetBEUI is good and efficient for small LAN and TCP/IP is suitable for larger networks such as WAN, why bothering to support IPX/SPX on Windows NT?" To answer this question, some background information is required.

Prior to Microsoft's launching of the Windows NT server, Novell occupied almost all the PC networking market. Therefore, Novell is the biggest competitor of Microsoft in that market segment. Novell's NetWare 3.11 and NetWare 3.12 (the version supports backup server) were the dominant network OS which were very popular 5 years ago. Not many people knew Windows NT at that time. However now, everybody is talking about NT, NT...

Compatibility is the most important feature, which attracts and keeps customers. Microsoft did very well on the compatibility of its product, Windows NT and Windows 95 can run most DOS applications. At that time, Windows NT was not as popular as it is now and most people only know NetWare and did not know Windows NT.

Microsoft strategy was to support NetWare in its Windows system. Microsoft would say, 'Let's switch to Windows NT, we can also support IPX/SPX'. After people switched to Windows NT, Microsoft would say, 'We have a more efficient protocol NetBEUI on the LAN environment, you don't need to use IPX/SPX any more'. Similarly, an interesting story is Windows NT 3.51, which supports FAT, HPFS as well as NTFS. FAT comes from Microsoft's former product DOS. HPFS comes from OS/2. While NTFS is the new file system, which supports features that are more advanced. If many features need to be supported, no doubt, it will decrease the performance.

When Microsoft saw OS/2 is no longer popular, it removed HPFS in its Windows NT 4.0 system. The possibility exists that Microsoft will someday remove IPX/SPX from its operating system.

As previously mentioned, protocol is like human language. If you installed the same protocol into two computers, it means these two computers can communicate. We'll discuss the 'client/server' concept in next section. For now, you need to remember, 'If same protocol is installed in two computers, they are guaranteed to communicate each other and run client/server applications.'

Please note only client/server applications are guaranteed supported if you only have same protocol installed. Other services such as file sharing or print server may not be supported. Take a network installation running Windows NT as the OS, and NWLink as the protocol. Such a system setup does not let it access the NetWare file or print server. To access files or printers on a NetWare server, a redirector such as CSNW (Client Service for NetWare) or GSNW (Gateway Service for NetWare) must be installed.

Note that CSNW can be installed on both Windows NT server and workstation while GSNW only can be installed in the Windows NT server. CSNW and GSNW can be used to let Windows NT system access NetWare resources. While if you want to NetWare client access Windows resources such as file and print services, File and Print Services for NetWare must be installed in Windows NT server. NWLink is useful if there are NetWare client/server applications running that use Sockets or NetBIOS over the IPX/SPX protocol.

NetBEUI is the fastest and most efficient of the above three protocols. However, it only can be used at the workgroup level and not at the enterprise network. This is because it is a non-routable protocol. Therefore, If the computers running NetBEUI must be connected using bridges while not routers. (We'll discuss bridge and router in later sections).

While NetBEUI is not routable, both IPX and IP are routable protocol. Because Internet connects millions of networks, only routable protocol such as TCP/IP can be used.

NetBIOS (Network Basic Input/Output System) defines a software interface and a naming convention. Please note NetBIOS is not a protocol. It only provides programming interface over a protocol. For example, NetBT (A session-layer network service) is NetBIOS over TCP/IP protocol.

If you only installed NWLink while not CSNW or GSNW on your computer, you can still run the NetBIOS based application on that computer.