Cisco CCNA: Local Area Networking
Unit 1. Introduction to LANs
In this course, you will learn many aspects of local area networks, or LANs. You will learn to identify LAN components and configurations. You will also learn about the three most common types of LANs: Ethernet, Token Rings, and FDDI. These LAN types differ in the way they encapsulate data, gain access to the media, and handle errors.
In this unit, you will learn about basic LAN components and standards. Then you will learn the standards of Ethernet, Token Ring, and FDDI, as well as how these LAN types operate.
After completing this unit, you should be able to:
- Identify the different LAN topologies
- Identify various cables used in LAN connections
- Recognize the organizations governing the different network types
- Distinguish between physical and logical addresses
- List the components of a MAC address
This unit provides information that is relevant to the following CCNA exam objectives:
- Define and describe the function of a MAC address
Topic 1.1: LAN Basics
*Defining a Local Area NetworkA Local Area Network (LAN) is a collection of computers in a limited geographical area that are connected in order to share resources. A LAN consists of nodes (workstations, servers, peripherals), their network cards, and the cables connecting everything together.
*Sharing and Communication
One of the many benefits of using LANs is that it allows users to share data. LANs also allow many different users to share expensive peripherals, such as printers, scanners, and plotters. One of the most popular LAN benefits is that they make communication faster and easier.
*Topologies and Cable
In this course you will learn about the different types of LANs, such as Ethernet, Token Ring, and Fiber Distributed Data Interface (FDDI). On the Physical layer of the OSI model, these types of LANs can be distinguished by their topologies and choice of cable.
Topic 1.1.1: LAN Topology
*LAN TopologyLAN topology comes in three basic physical forms: bus, ring, and star. Large networks may combine different LAN topologies into a more complex network.
*The Bus Topology
A bus is a series of computers in a linear arrangement. A terminator placed at each end-point of a bus network absorbs signals sent out by network devices so that the signals do not reflect back down the line. This is usually the easiest and least expensive topology to implement, and therefore very common.
*The Ring Topology
A ring is a series of computers connected in a ring arrangement. Each computer is connected to two adjacent computers to form a closed loop. A ring topology eliminates the need for terminators, but it is not easily administered.
*The Star Topology
A star is a group of computers connected to a central hub. This arrangement allows for efficient central administration and cable troubleshooting. The star topology also has the ability to mimic a ring topology in what is known as a logical ring.
*A Bus Star Topology
A cascading star (or bus star) is a number of stars connected to a bus backbone. Cascading stars are very common in Ethernet implementations and they give network administrators greater control of the network.
Topic 1.1.2: LAN Cabling
*Choosing CableThe choice of cable is a very important aspect in LAN operation. The most popular cables in use today are copper (coaxial, unshielded twisted pair, shielded twisted pair) and fiber optic. Each of these cable types has advantages and disadvantages, and network architects need to know which cables are best suited for specific tasks.
*Coaxial Cable
Coaxial cable has an inner conductor surrounded by insulation, a metal screen around the insulation, and an outer plastic cover. Coaxial cable has the advantages of carrying a large bandwidth, relative immunity to electromagnetic interference, the ability to carry signals over a significant distance, and the fact that many cable installers are familiar with it.
*Unshielded Twisted Pair Cable
Unshielded twisted pair cable (UTP) is lighter, thinner, more flexible, and easier to install than coaxial or fiber-optic cable. It is also inexpensive. It is the cable of choice for most locations, but is subject to interference from electrical devices. There are five categories or grades of UTP (Categories 1-5) with higher categories being better choices for data transmission.
*Shielded Twisted Pair Cable
Shielded twisted pair cable (STP) is a twisted pair cable surrounded by an RF-insulating material, making it immune to electrical interference. It is more expensive than UTP and must be grounded correctly to work properly.
*Fiber Optic Cable
Fiber optic cable transmits light pulses along a core fiber surrounded by cladding, which is a shield that reflects signals back to the fiber to reduce signal loss, and a protective cover. It has enormous bandwidth and can carry signals for extremely long distances, it is immune to electromagnetic interference, and it is more secure than copper cable. It is also expensive and can be difficult to install and maintain.
*Distance Constraints
There is a physical relationship, known as the distance constraint, between the length of a cable and the speed of data transmission. Faster transmission rates reduce the distance that data can travel on a given cable. A device known as a repeater may extend these distance constraints, but there is an upper limit to cable length.
*Using Repeaters
Repeaters receive a signal from an input port and transmit the signal to its output port. This amplifies the signal and increases the effective segment distance of a network. Thus, repeaters are considered a part of the overall cabling scheme of a network.
Topic 1.1.3: Types of LANs
*LAN ArchitectureFour types of LAN architecture in use today are Ethernets, Token Buses, Token Rings, and FDDIs. The standards for Ethernet, Token Bus, and Token Ring are governed by the IEEE (Institute of Electrical and Electronic Engineers) 802 committee. The IEEE 802 committee has five subcommittees covering the different types of LANs and LAN issues. FDDIs are governed by ANSI (American National Standards Institute).
*Governing the Logical Link Control Sublayer
IEEE 802.1 covers issues shared by all LANs. IEEE 802.2 governs the Logical Link Control sublayer of the OSI model. The other subcommittees each control standards for specific LAN types.
*Ethernet Standards
Ethernet standards are governed by IEEE 802.3. An Ethernet is a logical bus network with a backbone attached to the LAN nodes and with a terminator at each end. Ethernets may be physically configured in a star formation, but retain the characteristics of a logical bus. Ethernets control media access by a method known as CSMA/CD (Carrier Sense Multiple Access/Collision Detection).
*Token-Passing Networks
Token-passing networks use an electronic token to determine media access. These networks will be discussed more thoroughly in a later unit of this course.
*A Logical Ring of Computers
Token Ring networks use standards controlled by IEEE 802.5. A Token Ring is a logical ring of computers that take turns broadcasting. An electronic token is used to determine which node will broadcast. Token rings may be configured in a star formation, but retain the characteristics of a logical ring.
*Token Bus Standards
Token Bus standards are controlled by IEEE 802.4. Token Bus networks, developed by General Motors, are not very common. They were developed for computer-controlled manufacturing applications. Media access is controlled by the possession of an electronic token.
*Fiber Distributed Data Interface
FDDI, or Fiber Distributed Data Interface, is a token-passing network connected by fiber optic cables. These networks are very fast, with data rates of up to 100 Mbps.
FDDI LANs are configured as dual token rings for higher performance and also as backup in case of a failure in one ring. FDDI standards are governed by ANSI.
Topic 1.2: LAN Addressing
*Using AddressesAddresses are used in LANs to insure that information gets from the sending node to the receiving node. There are two types of addresses: physical addresses and logical addresses. These two address types operate on separate layers of the OSI model.
*Physical Addresses Are Permanent
Physical addresses (also known as link-layer, hardware, BIA, or burned-in addresses) are addresses that are an integrated part of the network card. They are burned into ROM by the manufacturers and will not change over time or when a LAN is reconfigured.
*Unique Physical Addresses
Physical addresses are unique for each network connection. They reside in what is known as a flat address space. A physical address is to a network connection as a social security number (SSN) is to an American. Just as no two people can have the same SSN, no two network cards can have the same physical address.
*Physical Addresses in the OSI Model
In the OSI model, physical addresses reside in the Media Access Control (MAC) sublayer of the Data Link layer. Therefore physical addresses are also known as MAC addresses.
Defining unique MAC addresses is the most important function of the MAC sublayer.
*MAC Addresses
Physical addresses, or MAC addresses, are only used within a local domain, which is a group of network devices using the same protocols and administered as a unit. Physical addresses are attached to a frame to deliver data to another network component within the local domain. To send data outside the local domain, a logical address is needed.
*Logical Addresses
Logical addresses (also called virtual addresses or Network-layer addresses) are used to transport data to a destination outside the local domain. Data traveling outside the local domain is attached to packets (or datagrams) containing the logical address of the originating node and the destination node.
*Following a Logical Address
Logical addresses are hierarchical, like mail addresses. Just as mail would have to travel to a city, a street, a building, and a mailbox to get to a person, a data packet may need to travel through hub A, bridge B, router C, and hub D to get to destination E.
*Resolving the Logical Address
When a packet arrives at the local domain of the destination, the logical address must be resolved to the physical address of the destination node. ARP, or Address Resolution Protocol, is a commonly used protocol that maps logical addresses to physical addresses. This will be covered in a later course in this series.
Question 1
Topic 1.3: The MAC Address
*The Hardware AddressThe MAC, or physical, address is the hardware address of a network node. These addresses are used to route data frames within a local domain. MAC addresses are 6 bytes (48-bits) long and are expressed in 12 hexadecimal digits. Hexadecimal numbers are covered in another course in this series.
*The Organizational Unique Identifier
The first 3 bytes (6 hexadecimal digits) of the MAC address contains the OUI (Organizational Unique Identifier) which is the manufacturer's identification or vendor code. OUIs are administered by the IEEE to make sure each manufacturer has a unique OUI.
*The Interface Serial Number
The last 3 bytes (6 hexadecimal digits) of the MAC address usually represent the interface serial number. There are no current standards for this part of the MAC address, and each vendor has its own method of determining these numbers.
*Transmitting a Frame
When a node is ready to transmit a data frame within the local domain, it places the MAC address of both the sending node and the receiving node onto the frame. The frame is then transmitted over the local network. The receiving node will process the frame when it recognizes its own MAC address on the frame.
Question 2
* Exercise 1
Try using the World Wide Web to find out more about local area networks.
| Step | Action |
|---|---|
| 1 | Use your browser to navigate to your favorite search engine (i.e., Yahoo!, LookSmart, Excite, etc.) |
| 2 | Search for terms you found in this unit. |
| 3 | Follow the links to find more information on the terms you selected. |
Topic 1.4: Unit 1 Summary
In this unit you learned how to identify the different LAN topologies and the different types of cable used to connect LAN components together. You also learned about different types of LANs and the agencies that govern the standards for each type.In addition, you learned to distinguish between physical and logical addresses. You learned about the MAC address, and how the MAC address uniquely identifies each component in a LAN.
In the next unit of this course, you will learn much more on Ethernet.
Unit 2. Ethernet
In this unit, you will learn about Ethernets. You will learn about the different types of Ethernet and how Ethernet networks package and transmit data. You will also learn how Ethernets control access to the media and handles transmission errors.
In addition, you will learn about the requirements and capabilities of half-duplex and full-duplex Ethernet operation.
After completing this unit, you should be able to:
- Identify the different Ethernet types
- Identify various cables used in Ethernet connections
- Distinguish between Ethernet and IEEE 802.3 frames
- List the methods of CSMA/CD
- Distinguish between half-duplex and full-duplex Ethernet
This unit provides information that is relevant to the following CCNA exam objectives:
- Describe full- and half-duplex Ethernet operation
Topic 2.1: Ethernet Basics
*Understanding EthernetEthernet is a type of LAN developed by Xerox in cooperation with Digital Equipment Corporation (DEC) and Intel. Ethernet LANs are very sturdy and reliable, and Ethernets are the most popular type of LAN in use today.
*Data Link Differences
Ethernet is almost identical to the IEEE 802.3 standard and are both governed by the IEEE 802.3 subcommittee. However, they have differences in the Data Link layer of the OSI model. IEEE 802.3 splits the Data Link layer into the MAC and LLC sublayers. Ethernet does not support LLC services, but conforms to the rest of the IEEE 802.3 standard.
*Ensuring Data Transfer
Ethernet and IEEE 802.3 use the CSMA/CD protocol to ensure proper data transfer. The CSMA/CD standard gives devices the ability to detect a data collision, which occurs when two network devices transmit at the same time and the data become garbled. After a collision is detected, a device will retransmit the message after a random delay time.
Topic 2.1.1: Ethernet Types
*Ethernet Bus TopologyEthernet and IEEE 802.3 have a bus topology and can have a signaling rate of 10, 100, or even 1000 Mbps. LAN nodes are connected to the bus to form the network. Terminators are placed at both ends of the bus to absorb signals, which prevents signal reflection. Standard names for Ethernets include the signal speed. For example, an Ethernet name beginning with 10Base would operate at 10 Mbps and one beginning with 100Base would operate at 100 Mbps.
*Thin Ethernet
10Base-2, or Thin Ethernet, uses 0.25-inch diameter coaxial cable. Each 10Base-2 cable segment may be up to 185 meters in length. Thin Ethernet cables are connected to individual nodes with the use of BNCs (British Naval Connectors) and BNC tees.
*Thick Ethernet
10Base-5, or Thick Ethernet, uses 0.4-inch diameter coaxial cable. Each segment on Thick Ethernet may be up to 500 meters in length. Thick Ethernet cables are connected to individual nodes with the use of transceivers and AUIs (attachment unit interfaces). Transceivers transmit and receive data to and from the network, and AUIs provide the connection for the cable to the node.
*10BaseT
10Base-T uses UTP cable up to 100 meters in length. UTP cable can only connect at the cable end points, and therefore connects only two nodes. For this reason, 10Base-T almost always uses a hub, which is a central connection point for all the nodes. The hub is also considered a node, with cable segments attached to all the other nodes.
10Base-FL is similar to 10Base-T, but uses fiber optic cable.
*Fast Ethernet
Fast Ethernet operates at 100 Mbps. Examples of Fast Ethernet are 100Base-TX, with UTP cables up to 100 meters in length, and 100Base-FX, with fiber optic cables up to 400 meters in length. There is also a Gigabit Ethernet, which operates at 1000 Mbps. Fast Ethernet and Gigabit Ethernet are covered in the next unit of this course.
Question 3
Question 4
Topic 2.1.2: Ethernet Interface
*Network Interface CardsEthernet network interface cards (NICs) allow each network device, or node, to communicate across the cable media. Each node is identified by its MAC address.
*NIC Connections
10Base-5 is connected to the bus with AUIs and transceivers. The AUI is a 15-pin socket located on the NIC.
A cable connects the AUI to the transceiver, which is attached to the bus. Some NICs have a built-in transceiver, in which case the AUI can be directly attached to the bus.
*Connecting 10Base-T
The interface for 10Base-T is different from those of 10Base-2 and 10Base-5. Since 10Base-T uses UTP cables with RJ-45 connectors, the network cable is plugged directly into the NIC. The other end of the cable is attached to the network hub.
*Ethernet Hubs
Ethernet hubs provide a central location for linking the nodes of the network. They also serve as an attachment site to link the local network to a larger network. This centralized wiring makes it much easier for administrators to troubleshoot cable problems. It is important to remember that even when an Ethernet uses a star topology, it still behaves like a bus in that all nodes receive all transmissions at about the same time.
Topic 2.2: Ethernet/802.3 Operation
*Packets to FramesEthernet nodes encapsulate data that needs to be sent over the network into a packet. When the packet is ready to be sent over the local network, the node adds source and destination MAC addresses to the packet. This process converts the packet into a frame.
*Transmitting Frames
When a node transmits a frame, the signal travels the entire network. Every node in the network receives and examines the frame. When the data reaches the endpoints of the bus, terminators absorb the signal to prevent reflection.
*Processing a Frame
When a node receives a frame, it checks for the destination MAC address in the frame. If that MAC address matches its own MAC address, the node will process the frame. If the MAC addresses do not match, the frame is discarded.
*Frame Errors
Frames are also discarded if they contain errors. An example of a frame error would be when a frame does not conform to size specifications. Frames must be between 64 and 1518 bytes in length. Frames less than 64 bytes in length are called runts. Frames greater than 1518 bytes are called giants. Both runts and giants are considered errors and are discarded by network nodes.
*Frames and the OSI Model
Frames operate on the Data Link layer of the OSI model. They must be moved from the Data Link layer to the Physical layer to be transmitted over the network. Ethernet and IEEE 802.3 frames have differences because IEEE 802.3 frames provide fields for LLC services and Ethernet does not.
*Transmission Time
The time required to move a frame between the Data Link layer to the Physical layer is called the transmission time. Each Ethernet bit has a window of 100 nanoseconds. A packet size of 64 bytes transmits in 51.2 microseconds. A packet of:
512 bytes transmits in 410 microseconds
1000 bytes transmits in 800 microseconds
1518 bytes transmits in 1214 microseconds.
Question 5
Topic 2.2.1: Ethernet Frames
*The PreambleEthernet frames begin with a preamble. The preamble is an eight-byte segment that provides clock synchronization and lets receiving stations know that a frame is coming. The preamble ends with a 10101011 bit pattern, called the Starting Delimiter, to indicate the arrival of data.
*Destination Address
The preamble is followed by the destination address (DA). This is the six-byte MAC address of the intended receiver of the frame. This address may include a single node (unicast), a group of nodes (multicast), or all nodes on the network (broadcast). The unicast address is the destination node's MAC address, the multicast address sets the first transmitted bit of the DA to a value of 1, and the broadcast address is all 1s (hexadecimal FFFF.FFFF.FFFF).
*Source Address
The DA is followed by the source address (SA). This is the six-byte MAC address of the originator of the frame. Since the originator of a frame is always a single node, the source address will always be unicast.
*Type Field
The SA is followed by a type field. This field is two bytes long and identifies the upper-layer protocol that will receive the data in the frame after the frame is received and processed by the lower-layer protocols.
*Data
The type field is followed by the actual data being transmitted. This data can be anywhere from 46-1500 bytes in length.
*Frame Check Sequence Field
Following the data is a frame check sequence (FCS) field. This field is four bytes long and contains the cyclical redundancy check (CRC) value and a remainder number. These values are used to check for errors in transmission. When a destination node receives a frame, it divides the length of the frame by the CRC value. The remainder should equal the remainder number in the FCS. If the remainders do not match, the frame is considered damaged, probably due to a collision.
Question 6
Question 7
Topic 2.2.2: IEEE 802.3 Frames
*IEEE 802.3 FramesThe first three sections of an IEEE 802.3 frame are identical to an Ethernet frame. An eight-byte preamble is followed by a six-byte DA, followed by a six-byte SA. The last section is the same as well: a four-byte FCS field.
*Length Field
After the source address, however, IEEE 802.3 frames contain a two-byte length field instead of a type field. The length field is used to indicate the length of the data (in bytes) that lie between the length field and the FCS field. The length field is then followed by either an 802.2 LLC (Logical Link Control) header or a SNAP (SubNetwork Access Protocol) header.
*802.2 LLC Header
An 802.2 LLC header consists of three segments of one byte each. They are the DSAP (Destination Service Access Point), SSAP (Source Service Access Point), and a control field. The DSAP and the SSAP are both pointers to memory buffers that tell the NIC where to put the information. This is very handy when multiple protocol stacks are running.
*SNAP Header
The SNAP header is designed to allow vendors to use their own protocols (called proprietary protocols) in the 802.2 LLC frame. In a SNAP header, the DSAP and SSAP are both set to the hexadecimal value of AA and are followed by a control field. This is then followed by the vendor OUI (3 bytes) and the Ethernet type for the frame (2 bytes).
Question 8
Question 9
Question 10
Question 11
Topic 2.2.3: CSMA/CD
*Collision DetectionCSMA/CD (Carrier Sense Multiple Access with Collision Detection) is an important aspect of Ethernet operation. It is a very reliable method to provide an orderly transmission of data. In CSMA/CD, nodes constantly monitor the network for transmissions. When a node is ready to transmit, CSMA/CD allows the node to determine whether other nodes are currently transmitting. If there are no transmissions on the network, the node will then proceed with the transmission.
*Jam Signals
Sometimes, however, two different nodes may begin transmitting at approximately the same time, causing a collision. When the transmitting node realizes a collision has occurred, it will send out a jam signal to alert the other nodes. All nodes then stop transmitting for a random time period, called the backoff time, before trying to retransmit.
*Retransmitting
If the backoff times of two nodes are too close to the same value, then subsequent collisions will occur and the process will repeat. The backoff time doubles after each collision until the 10th retry. After the 16th retry, a node will stop trying to transmit the data.
Topic 2.3: Half-Duplex versus Full-Duplex Ethernet
*Half-Duplex EthernetHalf-duplex Ethernet is the most popular of all LAN topologies. It uses a single channel to both transmit and receive data. Transmitting and receiving on the same channel can cause problems like collisions and loopbacks. Loopbacks occur when the sending node operates at a faster speed than the receiving node, and the signal loops back. Half-duplex Ethernet uses CSMA/CD and loopback detection to deal with collisions and loopbacks.
*Full-Duplex Ethernet
Full-duplex Ethernet, on the other hand, can transmit and receive simultaneously. This is accomplished by using two separate channels in a point-to-point connection from the transmitter of one device to the receiver of the receiving device. Full-duplex Ethernets do not share bandwidth, and are thus collision-free systems.
*Collision-Free
Since a full-duplex Ethernet is collision free, the CSMA/CD protocol does not have a function in this mode. The CSMA/CD protocol needs to be disabled in any full-duplex Ethernet implementation.
*Implementing Full-Duplex Ethernet
Full-duplex Ethernet can be used in 10Base-T, 100Base-T, 10Base-FL, and 100Base-FX. To implement a full-duplex Ethernet, requirements below must be met.
*Avoiding Problems
Problems will occur if half- and full-duplex modes are mixed. If one side of a connection is configured for full-duplex and the other side of the connection is configured for half-duplex, a bad FCS will occur. If the bad FCS has an integral number of octets, the error is called an FCS error. If the bad FCS has a non-integral number of octets, the error is called an alignment error.
Question 12
* Exercise 1
Try searching the World Wide Web for more information on Ethernets.
| Step | Action |
|---|---|
| 1 | Use your browser to navigate to your favorite search engine (i.e., Yahoo!, LookSmart, Excite, etc.) |
| 2 | Search for terms you found in this unit. |
| 3 | Follow the links to find out more information on the terms you selected. |
Topic 2.4: Unit 2 Summary
In this unit you learned how to identify the different Ethernet types and the various cables used in Ethernet connections. You also learned how to distinguish between Ethernet and IEEE 802.3 frames and how CSMA/CD works to control media access and errors. In addition, you learned how to distinguish between half-duplex and full-duplex Ethernet.In the next unit you will learn about Fast Ethernets.
Unit 3. Fast Ethernet
In this unit you will learn about the different types of Fast Ethernet. You will learn about the cables and distance limitations in Fast Ethernet. You will also learn how to calculate Fast Ethernet cable lengths when repeaters are used and when fiber optic cable is used to replace a copper cable segment.
Finally, you will learn about Gigabit Ethernet.
After completing this unit, you should be able to:
- List the different types of Fast Ethernet
- Identify the cables used for different types of Fast Ethernet
- List the guidelines and distance limitations of Fast Ethernet
- Calculate Fast Ethernet cable lengths when repeaters are used
- List the differences between Fast Ethernet and Gigabit Ethernet
This unit provides information that is relevant to the following CCNA exam objectives:
- Describe the features and benefits of Fast Ethernet
- Describe the guidelines and distance limitations of Fast Ethernet
Topic 3.1: What Is Fast Ethernet?
*Fast EthernetFast Ethernet is a form of Ethernet with data transfer rates much higher than the 10 Mbps rate of classic Ethernet. The rate at which Fast Ethernet transfers data is 100 Mbps, and is known as 100Base-T. 100Base-T was approved as the Fast Ethernet standard in 1995 by the IEEE and designated IEEE 802.3u.
*Fast Ethernet Collision Window
Most implementations of Fast Ethernet use the CSMA/CD LAN media access method, just like classic Ethernet. However, since the bandwidth is increased by a factor of ten, the collision window, which is the amount of time in which a collision may occur, is reduced to one tenth.
*Implementing Fast Ethernet
Implementations of Fast Ethernet include 100Base-T4, 100Base-TX, 100Base-FX, and 100VG-AnyLAN. 100Base-T4, 100Base-TX, and 100Base-FX all have MAC layers that are compatible with the IEEE 802.3 MAC layer. The MAC layer for 100VG-AnyLAN, however, is not compatible with the IEEE 802.3 standard because it uses a different method to control media access.
*100Base-T4
100Base-T4 varies from 10Base-T at the Physical layer. In 100Base-T4, the 100-Mbps data stream is divided into three 33-Mbps streams. The three streams created are then sent over three pairs of UTP wire. A fourth pair of UTP wire is used by the hub to send a collision signal to a workstation when a collision occurs.
*100Base-TX Protocols
The 100Base-TX is a true derivative of 10Base-T and both use the same protocols. Thus, 100Base-TX supports both half- and full-duplex modes. The 100-Mbps speed is achieved by sending the signal 10 times faster than 10Base-T. Signal strength is retained by allowing only two repeaters between nodes, as opposed to four repeaters between nodes in 10Base-T.
*100Base-FX
100Base-FX is similar to 100Base-TX, but uses fiber optic cable instead of copper wires. Segments are usually limited to 400 meters, but 62.5/125-micron fiber optic cables in full-duplex mode have no distance restrictions. Since 100Base-FX and 100Base-TX are similar in all but cable type, they are sometimes grouped together and designated as 100Base-X.
*100VB-AnyLAN
100VG-AnyLAN uses Category 3 or 5 UTP and utilizes aspects of both Ethernet and Token Ring. When a node is ready to transmit, it sends a request to the hub, which then grants access to the media depending on priority and network load. The hub for 100VG-AnyLAN must have at least one uplink port, and all other ports must be capable of being used as downlink ports. Token Rings are discussed in the next unit of this course.
Topic 3.1.1: Cabling for Fast Ethernet
*Cabling 100Base-TX100Base-TX uses two pairs of Category 5 UTP with RJ-45 connectors. This type of cable cancels electromagnetic interference by increasing the twist ratio of the wires. However, it is more expensive than Category 3 UTP.
The maximum length for a cable segment is 100 meters.
*Cabling 100Base-T4
100Base-T4 requires four pairs of Category 3, 4, or 5 UTP cables with RJ-45 connectors. Since the data stream is split over three of the four pairs, the full-duplex mode is not supported.
The maximum segment length is 100 meters.
*Cabling 100Base-FX
100Base-FX uses a two-strand multimode fiber optic cable. This cable may have a diameter of 50/125 or 62.5/125 microns and requires an SC, ST, or MIC fiber optic connector.
The maximum segment length is 400 meters.
*Cabling 100VG-AnyLAN
100VG-AnyLAN uses four pairs of Category 3 or 5 UTP wires with RJ-45 connectors. The maximum segment length is 600 meters for Category 3 UTP, and 900 meters for Category 5 UTP. If the hubs are in the same closet, the maximum lengths drop to 200 meters for Category 3 UTP and 300 meters for Category 5 UTP. Cascaded hubs (for a cascaded star topology) have maximum distances between hubs of 100 meters (Category 3 UTP) or 150 meters (Category 5 UTP).
*Avoiding Late Collisions
Sometimes a network exceeds the IEEE 802.3 limitations on cable lengths. Exceeding these specifications may result in late collisions. A late collision is a collision that is detected late in the data transmission. Late collisions are usually not a big problem, but they can lead to a slowdown in network traffic.
Question 13
Question 14
Topic 3.1.2: Guidelines
*Upgrading Classic EthernetOne of the advantages of Fast Ethernet is that it allows a gradual upgrade from classic Ethernet. There are a few issues that need to be addressed during this process. For example, classic Ethernet can have up to 4 hubs or repeaters between nodes. In Fast Ethernet, only one hub is allowed per collision domain, which is a group of nodes that share the same bandwidth. This limitation may be overcome by using switching hubs, which break up the collision domain.
*Identifying Potential Bottlenecks
When upgrading an Ethernet segment from 10Base-T to 100Base-T, potential bottlenecks should be identified and switches placed to alleviate the bottlenecks. All cables should be inspected to make sure they are adequate for the implementation. Also, the NICs should be checked to see if they are capable of 100 Mbps speed.
*Autonegotiation
Networks that combine 10Base-T and 100Base-T usually have an assortment of NICs with varying capabilities. This can be very confusing for network administrators, especially since both 10Base-T and 100Base-T use RJ-45 connectors. IEEE 802.3 provides for a method to help with this problem, called autonegotiation.
*Transmitting a Pulse
In autonegotiation, nodes and hubs transmit a pulse along their connection. The 10Base-T pulse is called an NLP (Normal Link Pulse). If a node or hub detects a NLP from its connection, it will only transmit at 10 Mbps along that connection. The 100Base-T pulse is called an FLP (Fast Link Pulse), and contains a code word describing its capabilities. The station and hub then compare capabilities and automatically exchange data at the highest common level.
*Advantages of Fast Ethernet
The advantages of upgrading from classic Ethernet to Fast Ethernet include improvements in bandwidth, throughput, and overall performance. There is also an easy migration from classic Ethernet to Fast Ethernet using existing cables and network equipment — and both networks use the reliable CSMA/CD.
*Disadvantages of Fast Ethernet
There are few disadvantages in upgrading to Fast Ethernet, but there is an increase in costs and the use of the same type of cables and connectors can be confusing. An upgrade to a 100VG-AnyLAN might also produce problems associated with the different media access methods of 100VG-AnyLAN and classic Ethernet.
Question 15
Topic 3.2: Repeaters in 100BaseT
*RepeatersThere are two different types of repeaters that can be used in Fast Ethernet networks, Class I and Class II. Class I repeaters limit networks to only one repeater between nodes, but have more functionality than Class II repeaters. Class II repeaters are faster than Class I repeaters and allow networks to use multiple repeaters between nodes.
*Class I Repeaters
Class I Fast Ethernet repeaters perform the normal function of repeating signals, but they are also capable of identifying corrupt frames, translating data for different Ethernet implementations, and performing error handling. Class I repeaters also partition off malfunctioning nodes, such as nodes that stop using CSMA/CD, send endless streams of bits (jabbers), or transmit frames with invalid preambles.
*Stacking Repeaters
A Fast Ethernet network with a repeater connecting two nodes, and each node at the maximum 100 meter cable limit, has a total cable distance of 200 meters. Stacking two Class II repeaters allows a total distance of 205 meters, including the cable connecting the two repeaters.
*Cable Distance
The cable distance is the longest possible path between nodes. For example, Node 1 is attached with 40 meters of cable to Repeater 1, Repeater 1 is attached to Repeater 2 with 16 meters of cable, and Repeater 2 is attached to Node 2 with 75 meters of cable. The total cable distance would be 40 + 16 + 75 = 131 meters.
*Maximum UTP Cable Distances
Some Class II repeaters, such as Cisco FastHub 300, exceed the IEEE 802.3u specifications. These repeaters allow longer cable distances, provided the 100 meter UTP segment limit is not exceeded. The following distances are maximum UTP cable distances using repeaters: 1 repeater any type — 200 m, 2 Class II repeaters — 205 m, 1 FastHub + 1 other Class II repeater — 214 m, 2 FastHubs — 223 m.
*Fiber Optic Cables
Even longer cable distances may be achieved if fiber optic cables are used in place of one of the UTP cable segments. When determining maximum cable distances, fiber optic segments must be converted to their UTP equivalents by multiplying the fiber optic length by 0.9. When converting a UTP length to a fiber optic equivalent, multiply by 1.11.
*Single Repeaters and Fiber Optic Cable
If fiber optic cable is used for one of the cable segments, the maximum cable distances (in UTP equivalents) using a single repeater in 100Base-T networks then become: 1 Class I repeater — 245 m, 1 Class II repeater — 287 m, 1 FastHub — 296 m
*Calculating Cable Distances
Here is an example that demonstrates how to calculate cable distances when both UTP and fiber optic cable are used:
A FastHub, connected to node A with 50 meters of UTP and to node B with 85 meters of UTP, can be connected to node C with fiber optic cable as far away as 234 meters. This is calculated by using 296 (longest FastHub distance using UTP and fiber optic), subtracting 85 (the longest length of UTP from FastHub to node A or B), and multiplying by 1.11 (to convert UTP to fiber optic).
Question 16
Question 17
Question 18
Question 19
Question 20
Question 21
Topic 3.3: Gigabit Ethernet
*Gigabit EthernetGigabit Ethernet, or 1000Base-X, is a relatively new Ethernet implementation with data transfer rates of 1000 Mbps (1 Gbps). The standards for 1000Base-X are governed by IEEE 802.3z and combines aspects of the IEEE 802.3 Ethernet standards with elements of the ANSI X3T11 Fibre Channel technology. The result is a fast network that is identical to Ethernet from the Data Link layer upward.
*Gigabit Ethernet Cabling
Gigabit Ethernet can transmit over different media. 1000Base-LX uses long wave lasers over single mode and multimode fiber optic cables, 1000Base-SX uses short wave lasers over multimode fiber optic cables, and 1000Base-CX uses up to 25 meters of STP. Still in development is 1000Base-T, which will use Category 5 or higher UTP up to 100 meters in length.
*Operating Modes
Gigabit Ethernet can operate in either half-duplex or full-duplex modes. In half-duplex mode the IEEE 802.3 CSMA/CD is used, and problems occur when frames are smaller than the slot time. The slot time is the unit of time in bits for Ethernet MAC to handle collisions. When a frame is smaller than the slot time, Gigabit Ethernet adds bits called carrier extensions to a frame to make it bigger than the slot time.
*Frame Bursting
Gigabit Ethernet also uses a technique called frame bursting in the half-duplex CSMA/CD environment. Frame bursting allows a transmitting node to prevent other nodes from transmitting until it has sent all of its frames. This is done by filling the idle time between frames with a stream of bits so that other nodes will see that the network is busy.
*Gigabit Ethernet Advantages
Gigabit Ethernet is expected to become a popular LAN alternative due to its large bandwidth, the use of the IEEE 802.3 Ethernet frame format, and backward compatibility with slower forms of Ethernet. Initial deployment of 1000Base-X will probably be as a fast backbone for 100 Mbps subnetworks.
Question 22
* Exercise 1
Try searching the World Wide Web for more information on terms discussed in this unit.
| Step | Action |
|---|---|
| 1 | Use your browser to navigate to your favorite search engine (i.e., Yahoo!, LookSmart, Excite, etc.) |
| 2 | Perform a search for terms like Fast Ethernet, repeaters, Gigabit Ethernet, or other terms you are interested in. |
| 3 | Follow the links to find out more about the terms you selected. |
Topic 3.4: Unit 3 Summary
In this unit you learned about the different types of Fast Ethernet, such as 100Base-T4, 100Base-TX, 100Base-FX, and 100VG-AnyLAN. You learned about the cabling requirements for these implementations as well as their guidelines and distance limitations of Fast Ethernet. In addition, you learned how to calculate Fast Ethernet cable lengths when repeaters are used and when fiber optic cable is used in place of a copper segment. You also learned about Gigabit Ethernet.In the next unit, you will learn about Token Rings and FDDI.
Unit 4. Token Ring and FDDI
In this unit you will learn about Token Rings. You will learn about Token Ring operation and the active monitor. You will also learn the components of Token Ring frames.
In addition, you will learn about FDDI networks. You will learn about FDDI operation and the reliability of the dual ring. You will also learn the components of FDDI frames.
After completing this unit, you should be able to:
- Identify a Token Ring network and its components
- Assemble a Token Ring frame
- Identify an FDDI network and its components
- Assemble a FDDI frame
This unit does not address any specific Cisco objectives. However, it does provide background information that is essential for the CCNA exam.
In the course index, questions about background information are indicated with the abbreviation BCK and a short description of the question subject matter.
Topic 4.1: Token Ring
*Token Ring NetworksThe Token Ring network was developed by IBM. IEEE 802.5 and IBM's Token Ring are nearly identical specifications, and now the terms are virtually synonymous. Token Rings are characterized by their ring topology and the use of an electronic token-passing method to determine media access.
*Token Ring Topology
The Token Ring topology is a logical ring but a physical star. Each node is connected to a central hub called an MSAU (multistation access unit) by STP or UTP cables.
*Multistation Access Unit
Each MSAU can connect up to eight nodes. More nodes may be connected by using stackable Token Ring switches instead of shared MSAUs. This would also allow increased performance and port density.
*Lobe Length
The length of the cable from a node to the MSAU is known as the lobe length. The maximum lobe length for UTP cables is 100 meters. For STP cables this length is 300 meters.
*NICs
Token Ring NICs support transfer rates of either 4 or 16 Mbps, and some Token Ring NICs can be configured for either speed. NICs operating at 4 Mbps cannot receive data from NICs operating at 16 Mbps. If these speeds are mixed on a single Token Ring network, the integrity of the ring is destroyed and the ring will not operate properly.
*Protocol
IEEE 802.5 protocol follows IEEE 802.3 in instances like MAC sublayer and physical layer services and it relies on the 802.2 LLC sublayer and upper-layer protocols for point-to-point services. However, IEEE 802.5 uses a token to determine LAN medium access, while IEEE 802.3 uses CSMA/CD.
Question 23
Question 24
Topic 4.1.1: Token Ring Operation
*Taking Turns TransmittingToken Rings provide a collision-free network where nodes take turns transmitting. Nodes receive signals from their NAUN (nearest active upstream neighbor) and repeat those signals to their downstream neighbors.
*Token Passing
A node can transmit only when it is in possession of a token, and then it can transmit only a single frame. The node appends the frame to be transmitted to the token and sends it to the next node. If the node has no data to send, it passes the token to the next node where the process is repeated.
Token passing is always the first step in data transmission in Token Ring networks.
*Transmitting a Frame
When a node transmits a frame, the frame travels the ring and is examined by each node in turn. When the destination node receives the frame and recognizes the destination address as its own MAC address, it copies the frame and then tags the original frame in the Frame Status field.
The original frame then continues along the ring until it reaches the originating node, which reads the Frame Status field to see if the frame was received by the destination node and then removes the frame from the ring.
*Early Token Release Feature
There is only one frame on the Token Ring at any given time unless the ETR (early token release) feature is used. The ETR allows a node to send the token onto the ring immediately after transmitting a frame.
The next node would then receive the token while the frame is traversing the ring, in which case it could be possible for that node to transmit a frame while the first frame was still on the ring.
*Ring Access Priority
Certain nodes may require greater access to the ring than other nodes. Ring access priority is accomplished by placing priority and reservation bits on the token.
Nodes must have a priority greater than or equal to that on the token in order to use the token. Nodes can reserve a token only if they have a priority greater than that of the transmitting node.
*Beaconing and Fault Management
Token Rings contain a fault management algorithm called beaconing. Beaconing begins when a node detects a fault, such as not receiving anything from its NAUN. The node then sends out a beacon frame to define the failure domain, which includes the node, its NAUN, and everything in between.
All nodes in the failure domain then undergo autoreconfiguration in an attempt to reconfigure the network around the failed areas. This process is not always successful, and manual reconfiguration is sometimes necessary.
Topic 4.1.2: Token Ring Frames
*Token Ring FramesToken Ring frames consist of a 1-byte Starting Delimiter, a 1-byte Access Control field, a 1-byte Frame Control field, two 6-byte MAC addresses for the destination and source, a variable-length RIF (Routing Information Field), the encapsulated data, a 4-byte FCS (Frame Check Sequence), a 1-byte Ending Delimiter, and a 1-byte Frame Status field.
*Delimiters
The Starting and Ending Delimiters indicate the beginning and end of the frame or token. The Frame Control field indicates the frame type, and the RIF is used to indicate a route through a network.
*Access Control Field
The Access Control field contains Priority bits (P) to indicate the priority of the frame or token, Reservation bits (R) to indicate the priority required for the next token to access the ring, a Token bit (T) to differentiate a token from a frame, and a Monitor bit (M) to determine if a frame has been on the ring for too long. The pattern for the Access Control field is PPPTMRRR.
*Frame Status Field
The Frame Status field contains an address recognition bit (A-bit) and a copied bit (C-bit). The originating node sets these bits to 0 before sending the frame onto the ring. The receiving node changes the A-bit to 1 if it recognizes its own address and the C-bit to 1 if it successfully copies the frame. When the frame returns to the originating node, the node examines the A-bit and C-bit to see if the frame was successfully delivered.
*Frame Status Field Combinations
There are four possible combinations in the Frame Status field. 00 means the destination was not found. 01 means data was copied but not acknowledged. 10 means data was acknowledged but not copied. 11 means the data was acknowledged and copied, or sent to another ring by a bridge. The Frame Status field is in the form ACrrACrr, where A equals the A-bit and C equals the C-bit.
Question 25
Topic 4.1.3: The Active Monitor
*Active MonitorThere is always one node within the ring that acts as the active monitor. One of the jobs of the active monitor is to check the monitor bit (M-bit) of each frame as it travels around the ring. If this bit is a 0, the active monitor will change it to 1 and then forward the frame. When the active monitor detects an M-bit of 1, it will not forward the frame, thus removing the frame from the ring. This process is how a frame is removed if the node that originated the frame shuts down before the frame returns to it.
*Generating a New Token
Another job of the active monitor is to generate a new token if the old token is lost. The active monitor checks the TRT (token rotation time) and times out if the TRT exceeds a threshold value set by the network administrator — which would mean that a token has been lost.
If a token is lost, the active monitor generates a new token and sends it along the ring.
*Claiming Active Monitor Status
There can be only one active monitor at any given time, but any node on the ring can become the active monitor. For example, if the node that is the current active monitor is shut down, the first node to detect the loss of the active monitor and claim active monitor status will become the new active monitor.
Question 26
Topic 4.2: FDDI
*Dual Ring ConfigurationFDDI, or Fiber Distributed Data Interface, is another popular network type. FDDI uses single mode or multimode fiber optic cables configured as a dual ring.
In many respects, FDDI is similar to the Token Ring network, but the dual ring configuration allows greater flexibility and backup capabilities and the fiber optic medium allows faster data transfer rates of 100 Mbps.
*Nodes and Distance
A FDDI LAN may have a maximum of 500 nodes with a maximum distance between nodes (lobe length) of 2 kilometers using multimode fiber optic cables. The maximum circumference of a FDDI ring is 100 kilometers. If single mode cables are used, the maximum distance between two nodes is increased to 60 kilometers.
*Fiber Optic Cable
The use of fiber optic cable makes FDDI a good network choice in instances where there is a large distance between nodes, or where a large bandwidth is needed. The immunity of fiber optic cable to electromagnetic interference also makes FDDI a good choice in hostile environments.
*Other Network Media
Twisted pair cables (UTP and STP) may also be used as network media. If these cables are used, then the network is designated as a CDDI (Copper Distributed Data Interface). FDDI and CDDI are very similar in most respects, but twisted pair cables limit the lobe length to about 100 meters.
Question 27
Topic 4.2.1: FDDI Connections
*ModesA mode is a ray of light entering a fiber optic cable at a certain angle. Multimode cables use LEDs (light emitting diodes) to generate their signals and allow multiple modes.
This causes modal dispersion, a phenomenon where the different modes enter the cable at different angles and arrive at the destination at different times. Modal dispersion limits the bandwidth and length of multimode cables.
*Single Mode Fiber Optic Cable
Single mode fiber optic cables use lasers to generate the signal and allow only a single mode. There is no problem with modal dispersion in single mode fiber optic cable, so this cable does not limit the bandwidth like multimode cable does. Single mode fiber optic cable provides better connectivity over larger distances than multimode cable.
*FDDI Topology
The topology of a FDDI network is a physical star but a logical ring. Fiber optic cables connect each node to a dual ring network hub, called a DAC (Dual Attached Concentrator).
The two rings are the primary ring (shown in red) and the secondary ring (shown in yellow), and data travels along these rings in opposite directions. Normally, the primary ring is used for data transmission while the secondary ring is idle. This allows for a backup in case the primary ring fails.
*Single Attachment Station
Each node is attached via the DAC to the primary ring of the network. A node with only one connection to the DAC is known as an SAS (Single Attachment Station). An SAS will not affect the operation of the FDDI ring if it is powered down or disconnected.
*Dual Attachment Stations
Some nodes have a double connection to the DAC and are attached to both the primary ring and the secondary ring. These nodes are known as DASs (Dual Attachment Stations).
A powered down or disconnected DAS will affect operation of the FDDI ring, but this can be overcome with the optical bypass switch located in the NIC. This switch uses mirrors to route the light signals through itself to maintain ring integrity when the DAS is disconnected or powered down.
*Dual Homing
A technique called dual homing allows for additional network reliability. In this technique, a dual-homed router or mainframe host has a single attachment to two different DACs on the FDDI dual ring. This provides for an active primary link as well as a backup connection to the FDDI network.
Question 28
Question 29
Topic 4.2.2: Operation
*FDDI OperationFDDI, like a Token Ring, uses tokens to determine which node may transmit data. Unlike Token Ring, however, FDDI nodes can attach new tokens to the end of a transmission. A downstream node can then attach its own frame to an existing frame and several frames may be on the ring at any given time.
*FDDI Frames
FDDI frames are very similar to Token Ring frames. However, they begin with a 8-byte Preamble, like Ethernet, for clock synchronization. The Preamble is followed by a 1-byte Start Delimiter, a 1-byte Frame Control field, the 6-byte Destination and Source addresses, the transmitted data, a 4-byte Frame Check Sequence, a 1-byte End Delimiter, and a 1-byte Frame Status field.
*Frame Transportation
A frame travels from the source node through the ring, being forwarded by each node in turn, until it gets back to the source node, which removes the frame from the ring by not forwarding it. A FDDI node must see its own MAC address in the frame before it removes the frame.
There are other fields in a frame before the Source Address, and these fields are forwarded before the source node recognizes its own MAC address and stops the forwarding process. This causes extra data to float around the ring. This extra data is periodically scrubbed to clear the network.
*FDDI Nodes and Active Monitors
All FDDI nodes are active monitors, and check for things like lost tokens, persistent frames (frames not removed from the ring by the source node), or faults in the ring. FDDI, like Token Ring, uses beacon frames to find faults. When a node does not receive tokens from its NAUN, the node will send out a beacon frames to find the fault and notify the rest of the ring. When the fault is found, the dual ring loops (or wraps) to form a single ring and thus bypass the problem.
Question 30
* Exercise 1
Try making a diagram of a FDDI network.
| Step | Action |
|---|---|
| 1 | Create a diagram of a FDDI network with two DACs and four SASs. |
| 2 | Now add two DASs. |
| 3 | Now add a dual-homed router. |
Topic 4.3: Unit 4 Summary
In this unit you learned about Token Ring and FDDI networks. You discovered how these networks operate and how they maintain integrity when nodes fail. You also learned the components of Token Ring and FDDI frames.You were presented with the basic principles of LAN operation in this course. You learned about LAN implementations such as Ethernets, Fast Ethernets, Token Rings, and FDDIs. You saw how each of these implementations packaged frames, granted access to the media, and controlled errors.
No comments:
Post a Comment