Hour 3: The Network Access Layer Page 1 of 10. Discuss how TCP/IP s Network Access layer relates to the OSI networking model

Transcription

1 Hour 3: The Network Access Layer Page 1 of 10 Hour 3: The Network Access Layer At the base of the TCP/IP protoco stack is the Network Access ayer, the coection of services and specifications that provide and manage access to the network hardware. In this hour you earn about the duties of the Network Access ayer and how the Network Access ayer reates to the OSI mode. This hour aso ooks at some common physica network technoogies you find in the Network Access ayer. At the competion of this hour, you be abe to Expain what the Network Access ayer is Discuss how TCP/IP s Network Access ayer reates to the OSI networking mode Expain what a network architecture is List the contents of an ethernet frame Identify the methods that ethernet, token ring, and FDDI use for controing access to the transmission medium Protocos and Hardware The Network Access ayer is the most mysterious and east uniform of TCP/IP s ayers. Basicay, the Network Access ayer manages a the services and functions necessary to prepare the data for the physica network. These responsibiities incude Interfacing with the computer s network adapter. Coordinating the data transmission with the conventions of the appropriate access method. You earn more about access methods ater in this hour. Formatting the data into a unit caed a frame and converting that frame into the stream of eectric or anaog puses that passes across the transmission medium. Checking for errors in incoming frames. Adding error-checking information to outgoing frames so that the receiving computer can check the frame for errors. Acknowedging receipt of data frames and resending frames if acknowedgment is not received. Of course, any formatting tasks performed on an outgoing frame must occur in reverse when the frame reaches its destination and is received by the computer to which it is addressed. The Network Access ayer defines the procedures for interfacing with the network hardware and fie://j:\macmiancomputerpubishing\chapters\jw036.htm 3/22/01

2 Hour 3: The Network Access Layer Page 2 of 10 accessing the transmission medium. Beow the surface of TCP/IP s Network Access ayer, you find an intricate interpay of hardware, software, and transmission-medium specifications. Unfortunatey, at east for the purposes of a concise description, there are many different types of physica networks that a have their own conventions, and any one of these physica networks can form the basis for the Network Access ayer. You earn about these physica network types ater in this hour. A few exampes incude Ethernet Token ring FDDI PPP (Point-to-Point Protoco, through a phone modem) Note - Not every networked computer is on a LAN. The network-access software may provide support for something other than a standard network adapter and a LAN cabe. One of the most common aternatives is a modem connection to a remote network, such as the connection you estabish when you dia in to an Internet service provider (ISP). Modem protoco standards such as Seria Line Internet Protoco (SLIP) and Point-to- Point Protoco (PPP) provide network access for the TCP/IP protoco stack through a modem connection. You earn more about these protocos in Hour 8, "Dia-Up TCP/IP." The good news is that the Network Access ayer is amost totay invisibe to the everyday user. The network adapter driver, couped with key ow-eve components of the operating system and protoco software, manages most of the tasks reegated to the Network Access ayer, and a few short configuration steps are usuay a that is required of a user. These steps are becoming simper with the improved pug-and-pay features of desktop operating systems. As you read through this hour, remember that the ogica, IP-stye addressing discussed in Hours 1, 2, 4, and 5 exist entirey in the software. The protoco system requires additiona services to deiver the data across a specific LAN system and up through the network adapter of a destination computer. These services are the purview of the Network Access ayer. Note - It is worth mentioning that the diversity, compexity, and invisibiity of the Network Access ayer has caused some authors to excude it from discussions of TCP/IP competey, asserting instead that the stack rests on LAN drivers beow the Internet Access ayer. This viewpoint has some merit, but the Network Access ayer reay is part of TCP/IP, and no discussion of the network-communication process is compete without it. The Network Access Layer and the OSI Mode fie://j:\macmiancomputerpubishing\chapters\jw036.htm 3/22/01

3 Hour 3: The Network Access Layer Page 3 of 10 As Hour 2, "How TCP/IP Works," mentioned, TCP/IP is officiay independent of the seven-ayer OSI networking mode, but the OSI mode is often used as a genera framework for understanding protoco systems. OSI terminoogy and concepts are particuary common in discussions of the Network Access ayer because the OSI mode provides additiona subdivisions to the broad category of network access. These subdivisions revea a bit more about the inner workings of this ayer. The OSI mode has been infuentia with computer networking vendors, and the recent trend toward mutiprotoco standards such as NDIS and ODI (discussed ater in this section) has accentuated the need for a common terminoogy that the OSI mode provides. As Figure 3.1 shows, the TCP/IP Network Access ayer roughy corresponds to the OSI Physica and Data Link ayers. The OSI Physica ayer is responsibe for turning the data frame into a stream of bits suitabe for the transmission medium. In other words, the OSI Physica ayer manages and synchronizes the eectrica or anaog puses that form the actua transmission. On the receiving end, the Physica ayer reassembes these puses into a data frame. Figure 3.1 OSI and the Network Access ayer. The OSI Data Link ayer performs two separate functions and is accordingy subdivided into the foowing two subayers: Media Access Contro (MAC) This subayer provides an interface with the network adapter. The network adapter driver, in fact, is often caed the MAC driver, and the hardware address burned into the card at the factory is often referred to as the MAC address. Logica Link Contro (LLC) This subayer performs error-checking functions for frames deivered over the subnet and manages inks between devices communicating on the subnet. Note - In rea network protoco impementations, the distinction between the ayers of TCP/IP and OSI systems has become further compicated by the deveopment of the Network Driver Interface Specification (NDIS) and Open Data-Link Interface (ODI) specification. NDIS (deveoped by Microsoft and 3Com Corp.) and ODI (deveoped by Appe and Nove) are designed to et a singe protoco stack (such as TCP/IP) use mutipe network adapters and to et a singe network adapter use mutipe upper-ayer protocos. This effectivey enabes the upper-ayer protocos to foat independenty of the network access system, which adds great functionaity to the network but aso adds compexity and makes it even more difficut to provide a systematic discussion of how the software components interreate at the ower ayers. Network Architecture In practice, oca area networks are not reay thought of in terms of protoco ayers but by what is caed LAN architecture or network architecture. (Sometimes a network architecture is referred to as a LAN type or a LAN technoogy.) A network architecture, such as ethernet, provides a bunde of specifications governing media access, physica addressing, and the interaction of the computers with fie://j:\macmiancomputerpubishing\chapters\jw036.htm 3/22/01

4 Hour 3: The Network Access Layer Page 4 of 10 the transmission medium. When you decide on a network architecture, you are in effect deciding on a design for the Network Access ayer. A network architecture is a design for the physica network and a coection of specifications defining communications on that physica network. The communication detais are dependent on the physica detais, so the specifications usuay come together as a compete package. These specifications incude considerations such as the foowing: Access method An access method is a set of rues defining how the computers wi share the transmission medium. To avoid data coisions, computers must foow these rues when they transmit data. Data frame format The IP-eve datagram from the Internet ayer is encapsuated in a data frame with a predefined format. The data encosed in the header must suppy the information necessary to deiver data on the physica network. You' earn more about data frames ater in this hour. Cabing type The type of cabe used for a network has an effect on certain other design parameters, such as the eectrica properties of the bitstream transmitted by the adapter. Cabing rues The protocos, cabe type, and eectrica properties of the transmission have an effect on the maximum and minimum engths for the cabe and for the cabe connector specifications. Detais such as cabe type and connector type are not the direct responsibiity of the Network Access ayer, but in order to design the software components of the Network Access ayer, deveopers must assume a specific set of characteristics for the physica network. Thus, the network access software must come with a specific hardware design. Physica Addressing As you earned in Part I, "TCP/IP Basics," the Network Access ayer is necessary in order to reate the ogica IP address, which is configured through the protoco software with the actua permanent physica address of the network adapter. The physica address is burned into the card at the factory. Data frames sent across the LAN must use this physica address to identify the source and destination adapters, but the engthy physica address (48 bits in the case of ethernet) is so unfriendy that it is impractica for peope to use. Aso, encoding the physica address at higher protoco eves compromises the fexibe moduar architecture of TCP/IP, which requires that the upper ayers remain independent of physica detais. TCP/IP uses the Address Resoution Protoco (ARP) and Reverse Address Resoution Protoco (RARP) to reate IP addresses to the physica addresses of the network adapters on the oca network. ARP and RARP provide a ink between the ogica IP addresses seen by the user and the (effectivey invisibe) hardware addresses used on the LAN. You' earn about ARP and RARP in Hour 4, "The Internet Layer." Anatomy of a Frame The Network Access ayer software accepts a datagram from the Internet ayer and converts that data fie://j:\macmiancomputerpubishing\chapters\jw036.htm 3/22/01

5 Hour 3: The Network Access Layer Page 5 of 10 to a form that is consistent with the specifications of the physica network (see Figure 3.2). Because many forms of physica networks exist, there are many formats for data at the Network Access ayer, and it woud not be easy or usefu to describe a these formats in detai. Figure 3.2 The Network Access ayer formats data for the physica network. As an exampe of what happens to the data at the Network Access ayer, consider the case of ethernet, the most common of the LAN architectures. When the ethernet software receives a datagram from the Internet ayer, it performs the foowing steps: 1. Breaks IP ayer data into smaer chunks, if necessary, which wi be sent in the data fied of the ethernet frames. The tota size of the ethernet frame must be between 64 bytes and 1,518 bytes (not incuding the preambe). 2. Packages the chunks of data into frames. Each frame incudes data as we as other information that the network adapters on the ethernet need in order to process the frame. An IEEE ethernet frame incudes the foowing: Preambe: A sequence of bits used to mark the beginning of the frame (8 bytes, the ast of which is the 1-byte Start Frame Deimiter). Recipient address: The 6-byte (48-bit) physica address of the network adapter that is to receive the frame. Source address: The 6-byte (48-bit) physica address of the network adapter that is sending the frame. Length: A 2-byte (16-bit) fied indicating the size of the data fied. Data: The data that is transmitted with the frame. Frame Check Sequence (FCS): A 4-byte (32-bit) checksum vaue for the frame. The FCS is a common means of verifying data transmissions. The sending computer cacuates a Cycica Redundancy Check (CRC) vaue for the frame and encodes the CRC vaue in the frame. The receiving computer then recacuates the CRC and checks the FCS fied to see if the vaues match. If the vaues don t match, some data was ost or changed during transmission. 3. Passes the data frame to ower-eve components corresponding to OSI s physica ayer, which wi convert the frame into a bitstream and send it over the transmission medium. The other network adapters on the ethernet receive the frame and check the destination address. If the destination address matches the address of the network adapter, the adapter software processes the incoming frame and passes the data to higher ayers of the protoco stack. IEEE is not the ony ethernet standard. The Ethernet II standard, used by some vendors, has a sighty different frame format. fie://j:\macmiancomputerpubishing\chapters\jw036.htm 3/22/01

6 Hour 3: The Network Access Layer Page 6 of 10 LAN Technoogies The most common network architectures are the foowing: Ethernet, incuding variants such as the foowing: 10BASE-2 (an ethernet standard using thin coaxia cabe) 10BASE-5 (an ethernet standard using thick coaxia cabe) 10BASE-T (an ethernet standard using twisted-pair cabe in a star configuration) 100BASE-TX (a standard simiar to 10BASE-T with faster transmissions speed (100Mbps) Token ring Note - The Institute of Eectrica and Eectronic Engineers (IEEE) has produced a set of standards for LAN architectures. Athough token ring and ethernet were both created before the IEEE standards, the IEEE specifications for IEEE (ethernet) and IEEE (token ring) now provide vendor-neutra standards for these important LAN technoogies. The foowing sections wi examine ethernet and token ring in greater detai, aong with another LAN technoogy: FDDI. Ethernet Ethernet and its newer sibing Fast Ethernet, are the LAN technoogies most commony used today. Ethernet has become popuar because of its modest price; Ethernet cabe is inexpensive and easiy instaed. Ethernet network adapters and Ethernet hardware components are aso reativey inexpensive. On ethernet networks, a computers share a common transmission medium. Ethernet uses an access method caed Carrier Sense Mutipe Access with Coision Detect (CSMA/CD) for determining when a computer is free to transmit data on to the access medium. Using CSMA/CD, a computers monitor the transmission medium and wait unti the ine is avaiabe before transmitting. If two computers try to transmit at the same time, a coision occurs. The computers then stop, wait for a random time interva, and attempt to transmit again. CSMA/CD can be compared to the protoco foowed by a room fu of poite peope. Someone who wants to speak first istens to determine if anybody ese is currenty speaking (this is the Carrier Sense). If two peope start speaking at the same moment, then both peope wi detect this, stop speaking, and wait before speaking again (this is Coision Detect). Traditiona ethernet works we under ight-to-moderate use but suffers from high coision rates fie://j:\macmiancomputerpubishing\chapters\jw036.htm 3/22/01

7 Hour 3: The Network Access Layer Page 7 of 10 under heavy use. Some of the newer ethernet variants, which may incude inteigent hubs or switches, support higher traffic eves. You earn more about hubs and switches in Hour 9, "Network Hardware." Ethernet is capabe of using a variety of media. Ethernet networks typicay operate at baseband speeds of either 10Mbps or 100Mbps. 1000Mbps (Gigabit) Ethernet systems are now avaiabe and may soon be common. Tabe 3.1 ists terms used to identify cabing media, speeds, and maximum distances. Wireess ethernet is aso becoming popuar. 10BASE-2 and 10BASE-5 coaxia ethernet networks were once very common. Figure 3.3 shows a coaxia 10BASE-2 network. Note that the computers are attached to a singe cabe that acts as the shared transmission medium. In recent years, hub-based ethernet variants such as 10BASE-T (see Figure 3.4) have become vasty more popuar. On a 10BASE-T network, the computers are attached to a centra hub. 10BASE-2 and 10BASE-T may appear to be dissimiar, but internay they are both sti ethernet. Tabe 3.1 Ethernet Media Technoogy Technoogy Name Media Type Operating Speed Maximum Distance 10BASE-2 Thin coax 10 megabits 185 meters 10BASE-5 Thick coax 10 megabits 500 meters 10BASE-T CAT3 or CAT5 10 megabits 100 meters UTP 10BASE-F Fiber optic 10 megabits 2,000 meters 100BASE-TX CAT 5 UTP or STP 100 megabits 100 meters 100BASE-FX Fiber optic 100 megabits 2,000 meters Figure 3.3 A 10BASE-2 coaxia ethernet network. Figure 3.4 A 10BASE-T hub-based ethernet network. Token Ring Token ring technoogy uses a competey different concept for aowing network adapters to transmit data on the media. This access method is known as token passing. Under the token passing access method, the computers on the LAN are connected so that data is passed around the network in a ogica ring (see Figure 3.5). The token ring configuration cas for the computers to be wired to a centra hub caed a MAU or MSAU. Figure 3.5 may not ook ike a ring, but the MSAU is wired so that the data passes from one computer to the next in a circuar motion. The computers pass a packet of data caed a token around the network. Ony the computer that hods the token can transmit a message on to the ring. fie://j:\macmiancomputerpubishing\chapters\jw036.htm 3/22/01

8 Hour 3: The Network Access Layer Page 8 of 10 Figure 3.5 A token ring. Token ring is technicay more sophisticated than ethernet, and it incudes a number of buit-in diagnosis and correction mechanisms that can hep troubeshoot network probems. Aso, because data is transmitted in a more ordery fashion, token ring does not suffer as bady under heavy data traffic. Amost everything about token ring is more expensive than ethernet by comparison the cabe, the network adapter cards, and the other components as we. Token ring typicay operates at either 4Mbps or 16Mbps. It is aso avaiabe at 100Mbps. FDDI Fiber Distributed Data Interface (FDDI) is an expensive LAN technoogy that empoys a pair of fiberoptic rings. One ring is considered primary and the second ring is principay there to repair the primary ring in the event of a breakdown. FDDI uses a token passing access method simiar to token ring. Like token ring, FDDI aso has error-detection and correction capabiities. In a normay operating FDDI ring, the token passes by each machine every so often. If the token is not seen within the maximum amount of time that it takes to circuate the argest ring, it indicates a probem has occurred such as a broken cabe. Fiber-optic cabe such as the cabe used with FDDI can support very arge voumes of data over arge distances. Other Network-Access Technoogies LAN technoogies such as ethernet are common throughout the mechanized word, but there are many other ways to connect computers. Any networking technoogy must have some means of preparing data for the physica network and, therefore, any TCP/IP technoogy must have a Network Access ayer. As was aready mentioned, a modem is another means of supporting a network connection. You' earn more about modems in Hour 8. Wide area network (WAN) technoogies support connections that operate over greater distances but often at sower transmission rates. WAN connections require speciaized hardware that, as you might guess, requires speciaized software at the Network Access ayer. Summary In this hour you earned about the Network Access ayer, the most diverse and arguaby the most compex ayer in the TCP/IP protoco stack. The Network Access ayer defines the procedures for interfacing with the network hardware and accessing the transmission medium. There are many types of LAN architectures and therefore many different forms that the Network Access ayer can take. This hour aso described the contents of an ethernet frame and briefy discussed ethernet, token ring, and FDDI. Q&A fie://j:\macmiancomputerpubishing\chapters\jw036.htm 3/22/01

9 Hour 3: The Network Access Layer Page 9 of 10 Q. What types of services are defined at the Network Access ayer? A. The Network Access ayer incudes services and specifications that manage the process of accessing the physica network. Q. Which OSI ayers correspond to the TCP/IP network Access ayer? A. The Network Access ayer roughy corresponds with the OSI Data Link ayer and Physica ayer. Q. What are the two most common LAN architectures? A. The most common LAN architectures are ethernet, with its severa cabing variants, and token ring. Q. What is CSMA/CD? A. CSMA/CD is Carrier Sense Mutipe Access with Coision Detect, a network-access method used by ethernet. Under CSMA/CD, the computers on a network wait for a moment to transmit and, if two computers attempt to transmit at once, they both stop, wait for a random interva, and transmit again. Q. What is token passing? A. Token passing is a network-access method used by token ring and FDDI. A packet of data caed a token circuates around the network. Ony the computer with the token can transmit data. Key Terms Review the foowing ist of key terms: Access method A procedure for reguating access to the transmission medium. CRC (Cycica Redundancy Check) A checksum cacuation used to verify the contents of a data frame. CSMA/CD The network access method used by ethernet. Data frame A package of data transmitted over an ethernet network. Data Link ayer The second ayer of the OSI mode. Ethernet A very popuar LAN architecture, using the CSMA/CD network-access method. FDDI A token passing network architecture using fiber-optic cabe. fie://j:\macmiancomputerpubishing\chapters\jw036.htm 3/22/01

10 Hour 3: The Network Access Layer Page 10 of 10 Logica Link Contro subayer A subayer of OSI's Data Link ayer that is responsibe for error checking and managing inks between devices on the subnet. Media Access Contro subayer A subayer of OSI's Data Link ayer that is responsibe for the interface with the network adapter. Network architecture A compete specification for a physica network, incuding specifications for access method, data frame, and network cabing. Physica address A permanent network address, burned into the adapter card by the manufacturer, that is used to deiver data across the physica network. Physica ayer The first OSI ayer, responsibe for transating the data frame into a bitstream suitabe for the transmission medium. Preambe A series of bits marking the beginning of a data frame transmission. Token passing The network access method used by token ring. Token ring A LAN architecture featuring a ring topoogy and a token-passing network access method. Copyright Pearson Education. A rights reserved. fie://j:\macmiancomputerpubishing\chapters\jw036.htm 3/22/01