CCNA 1 and 2 Companion Guide, Revised (Cisco Networking Academy Program) part 30 potx

History and Evolution of Ethernet 259
source device sends data out on a network, the data carries the MAC address of its
intended destination. As this data propagates along the network medium, the NIC in
each device on the network checks to see if its MAC address matches the physical des-
tination address carried by the data frame. If there is no match, the NIC discards the
data frame.
If there is a match, the NIC verifies the destination address in the frame header to
determine whether the packet is properly addressed. When the data passes its destina-
tion station, the NIC for that station makes a copy, takes the data out of the envelope,
and gives the data to the computer to be processed by upper-layer protocols such as IP
and TCP.
Framing in General
Encoded bit streams on physical media represent a tremendous technological accom-
plishment, but they alone are not enough to make communication happen. Framing
helps obtain essential information that could not otherwise be obtained with coded bit
streams alone. Examples of such information are listed here:
■ Which computers are communicating with one another
■ When communication between individual computers begins and when it terminates
■ Recognition of errors that occur during the communication
■ Whose turn it is to “talk” in a computer “conversation”
■ Where the data is located within the frame
When you have a way to identify computers, you can move on to framing. Framing is
the Layer 2 encapsulation process. A frame is the Layer 2 protocol data unit. Figure 5-5
illustrates the ideas of bits and frames.
When you work with bits, the most accurate diagram you can use to visualize them is
a graph showing voltage versus time. However, because you usually deal with larger
units of data and addressing and control information, this type of graph could become
ridiculously large and confusing. Another type of diagram you could use is the frame
format diagram, which is based on voltage versus time graphs. You read them from
left to right, just like an oscilloscope graph. The frame format diagram shows different
groupings of bits (fields); certain functions are associated with the different fields.

1102.book Page 259 Tuesday, May 20, 2003 2:53 PM
260 Chapter 5: Ethernet Fundamentals
Figure 5-5 From Bits to Frame
Many different types of frames are described by various standards. A single generic
frame has sections called fields, and each field is composed of bytes (see Figure 5-6).
The names of the fields commonly found in a data link layer frame are as follows:
■ Frame Start field
■ Address field
■ Length/Type/Control field
■ Data field
■ Frame Check Sequence (FCS) field
Figure 5-6 A Generic Frame Format
The sections that follow describe the different frame fields in more detail.
Field Names
ABC D E
Start
Frame
Field
Address
Field
Length/
Type/
Control
Field
Data
Field
FCS
Field
1102.book Page 260 Tuesday, May 20, 2003 2:53 PM
History and Evolution of Ethernet 261

Frame Start Field
When computers are connected to a physical medium, there must be a way for them to
grab the attention of other computers to broadcast the message, as in, “Here comes a
frame!” Various technologies have different ways of doing this process, but, regardless
of technology, all frames have a beginning signaling sequence of bytes.
Address Field
All frames contain identification information, such as the address of the source com-
puter (MAC address) and the address of the destination computer (MAC address).
Length and Type Fields
Most frames have some specialized fields. In some technologies, a Length field specifies
the exact length of a frame. Some have a Type field, which specifies the Layer 3 proto-
col making the sending request. There is also a set of technologies in which no such
fields are used.
Data Field
The reason for sending frames is to get higher-layer data, ultimately the user applica-
tion data, from the source computer to the destination computer. The data package
that you want to deliver includes the message that you want to send (the data). Pad-
ding bytes are added sometimes so that the frames have a minimum length. Logical
Link Control (LLC) bytes also are included with the data field in the IEEE standard
frames. Remember that the LLC sublayer adds control information to the network
protocol data, a Layer 3 packet, to help deliver that packet to its destination. Layer 2
communicates with upper layers through LLC.
Frame Check Sequence Field
All frames (and the bits, bytes, and fields contained within them) are susceptible to
errors from a variety of sources. The Frame Check Sequence (FCS) field contains a
number based on the data in the frame; this number is calculated by the source com-
puter. When the destination computer receives the frame, it recalculates the FCS num-
ber and compares it with the FCS number included in the frame. If the two numbers
are different, an error is assumed and the frame is discarded.
The Frame Check Sequence number normally is calculated through the use of a cyclic

redundancy check (CRC), which performs polynomial calculations on the data.
1102.book Page 261 Tuesday, May 20, 2003 2:53 PM
262 Chapter 5: Ethernet Fundamentals
Ethernet Frame Structure
At the MAC sublayer, the frame structure is nearly identical for all speeds of Ethernet
(10/100/1,000/10,000 Mbps). Half-duplex Gigabit Ethernet 1000BASE-T and the “W”
versions of 10-Gb Ethernet have certain timing issues that require minor differences in
how the interframe spacing is handled by the MAC sublayer, but these are otherwise
the same as the other speeds. However, at the physical layer, almost all versions of
Ethernet are substantially different from one another, and each speed has its own set
of architecture design rules.
IEEE 802.3 Ethernet Frame
In addition to the 802.2 frame type discussed previously, there is a simpler IEEE 802.3
frame type, which was the first developed by the IEEE. As with 802.2, however, it is
not used widely in today’s Ethernet LANs. Figure 5-7 shows the basic IEEE 802.3
Ethernet frame format.
Figure 5-7 IEEE 802.3 Ethernet Frame Structure
Table 5-2 lists the octet number and name for each 802.3 Ethernet frame field.
Table 5-2 IEEE 802.3 Ethernet Frame Fields
Octets in Each Frame Field Frame Field
7 Preamble
1 Start Frame Delimiter (SFD)
6 Destination MAC Address
6 Source MAC Address
2 Length/Type Field (Length if less than 0600 in
hexadecimal—otherwise, protocol Type)
46 to 1500 Data and Pad
4 Frame Check Sequence (CRC Checksum)
1102.book Page 262 Tuesday, May 20, 2003 2:53 PM
History and Evolution of Ethernet 263

Ethernet II Frame
In the DIX version of Ethernet that was developed before the adoption of the IEEE
802.3 version of Ethernet, the Preamble and Start Frame Delimiter (SFD) fields were
combined into a single field, although the binary pattern was identical. The field labeled
Length/Type was listed only as Length in the early IEEE versions and only as Type in
the DIX version. These two uses of the field officially were combined in a later IEEE
version because both uses of the field were common throughout the industry. The early
DIX Ethernet frame format, also know as Ethernet Version2 or Ethernet II, is the most
commonly used frame type with TCP/IP–based Ethernet LANs. Figure 5-8 illustrates
the Ethernet II frame format.
Figure 5-8 Ethernet II Frame Format
Table 5-3 lists the octet number and name for each Ethernet II frame field.
As indicated in Table 5-3, use of the Ethernet II Type field is incorporated into the cur-
rent 802.3 frame definition. Upon receipt, a station must determine which higher-layer
protocol is present in an incoming frame. This first is attempted by examining the
Length/Type field. If the two-octet value is equal to or greater than 600 hex, the frame
is interpreted according to the Ethernet II type code indicated. If it is less than 600 hex,
the frame is interpreted as an 802.3 frame and the length of the frame is indicated.
Further investigation is required to determine how to proceed. To proceed from here,
the first four octets of the 802.3 Data field are examined. The value found in those first
Table 5-3 Ethernet II Frame Fields
Octets in Each Frame Field Frame Field
8 Preamble (ending in pattern 10101011, the 802.3 SFD)
6 Destination MAC Address
6 Source MAC Address
2 Type Field (46 to 1500 data—if less than 46 octets, a
pad must be added to the end)
46 to 1500 Data and Pad
4 Frame Check Sequence (CRC Checksum)
1102.book Page 263 Tuesday, May 20, 2003 2:53 PM

264 Chapter 5: Ethernet Fundamentals
four octets usually is checked for two unique values; if they are not present, the frame
is assumed to be an 802.2 Logical Link Control (LLC) sublayer encapsulation and is
decoded according to the 802.2 LLC encapsulation indicated. One of the two values
tested for is AAAA in hexadecimal, which indicates an 802.2/802 (Subnetwork Access
Protocol [SNAP]) encapsulation. The other value tested for is FFFF in hexadecimal,
which might indicate an old Novell Internetwork Packet Exchange (IPX) “raw”
encapsulation.
Ethernet Frame Fields
The following list shows most of the fields permitted or required in an 802.3 Ethernet
Frame. Refer back to Figure 5-8 for an illustration of the 802.3 Ethernet frame:
■ Preamble—This field contains an alternating pattern of 1s and 0s that was used
for timing synchronization in the asynchronous 10-Mbps and slower implemen-
tations of Ethernet. Faster versions of Ethernet are synchronous, so this timing
information is redundant but is retained for compatibility. The preamble is seven
octets in length and is represented by the following binary pattern:
10101010 10101010 10101010 10101010 10101010 10101010 10101010
■ Start Frame Delimiter (SFD)—This a one-octet field that marks the end of the
timing information. It is represented by the binary pattern 10101011. In the
early DIX form of Ethernet, this octet was the last in the eight-octet preamble.
Although the old DIX Ethernet described the first eight octets differently than the
IEEE Ethernet, the pattern and usage is identical. Also, the timing information
represented by the preamble and SFD is discarded and is not counted toward the
minimum and maximum frame sizes.
■ Destination Address—This field contains the six-octet MAC destination address.
The destination address can be a unicast (single node), multicast (group of nodes),
or broadcast address (all nodes).
■ Source Address—This field contains the six-octet MAC source address. The source
address is supposed to be only the unicast address of the transmitting Ethernet
station. However, there are an increasing number of virtual protocols in use, which

use and sometimes share a specific source MAC address to identify the virtual
entity.
In the early Ethernet specifications, MAC addresses were optionally two or six
octets, as long as the size was constant throughout the broadcast domain. Two-
octet addressing first was excluded explicitly in paragraph 3.2.3 in the 1998 version
of 802.3 and no longer is supported in 802.3 Ethernet.
1102.book Page 264 Tuesday, May 20, 2003 2:53 PM
History and Evolution of Ethernet 265
■ Length/Type—If the value is less than 1536 decimal (0600 hexadecimal), value
indicates Length. The Length interpretation is used where the LLC layer provides
the protocol identification.
— Type (Ethernet)—The Type specifies the upper-layer protocol to receive the
data after Ethernet processing is complete.
— Length (IEEE 802.3)—The Length indicates the number of bytes of data
that follows this field. If the value is equal to or greater than 1536 decimal
(0600 hexadecimal), the value indicates Type, and the contents of the Data
field are decoded per the protocol indicated. A list of common Ethertype
protocols is found in RFC 1700, beginning around page 168.
■ Data and Pad—This field can be of any length that does not cause the frame to
exceed the maximum frame size. The maximum transmission unit (MTU) for
Ethernet is 1500 octets, so the data should not exceed that size. The content of
this field is unspecified. An unspecified pad is inserted immediately after the user
data when there is not enough user data for the frame to meet the minimum
frame length.
The frame structure figures depict the Data field as being between 46 and 1500
octets. In fact, Ethernet does not specify this. The frame is required to be not less
than 64 octets or more than 1518 octets, without actually specifying the size of
the data field. This left the user to calculate the size of the data field by subtracting
all the other fields from the frame size. If the currently required six octet MAC
addresses are used, the data field size will be between 46 (padded, if necessary)

and 1500 octets.
— Data (IEEE 802.3)—After physical layer and data link layer processing is
complete, the data is sent to an upper-layer protocol, which must be defined
within the data portion of the frame. If data in the frame is insufficient
to fill the frame to its minimum 64-byte size, padding bytes are inserted to
ensure at least a 64-byte frame.
■ Frame Check Sequence (FCS)—This sequence contains a 4-byte CRC value that
is created by the sending device and is recalculated by the receiving device to
check for damaged frames. The mathematical result of a cyclic redundancy check
(CRC) algorithm is placed in this four-octet field. The sending station calculates
the checksum for the transmitted frame, and the resulting four-octet value is
appended following the Data/Pad. Receiving station(s) perform the same calcula-
tion and compare the new checksum against the checksum found at the end of
the transmitted frame. If the two match, the frame is good. The fields used in the
calculation include everything from the beginning of the destination address to the
end of the Data/Pad as shown in Figure 5-7. The preamble, shown in Figure 5-8,
1102.book Page 265 Tuesday, May 20, 2003 2:53 PM
266 Chapter 5: Ethernet Fundamentals
illustrates that SFD and Extension fields are not included in the calculation. The
FCS is the only Ethernet field transmitted in noncanonical order (MSB first).
Because the corruption of a single bit anywhere from the beginning of the desti-
nation address through the end of the FCS field causes the checksum to be differ-
ent, the coverage of the FCS includes itself. It is not possible to distinguish between
corruption of the FCS itself and corruption of any preceding field used in the
calculation.
Ethernet Operation
When multiple stations (nodes) must access physical media and other networking devices,
various media access control strategies are used. This section briefly reviews the access-
control strategies and focuses on Ethernet access control method—CSMA/CD.
It should be noted that although CSMA/CD has immense historical importance and

practical importance in original Ethernet, it is diminishing somewhat in implementa-
tion for two reasons:
■ When four-pair UTP is used, separate wire pairs for transmission (Tx) and recep-
tion (Rx) exist making, copper UTP potentially free from collisions and capable
of full-duplex operation, depending on whether it is deployed in a shared (hub)
or switched environment.
■ Similar logic applies to optical fiber links, where separate optical paths—a trans-
mission fiber and an reception fiber—are used.
One new form of Ethernet—1000BASE-TX, Gigabit Ethernet over copper—uses all
four wire pairs simultaneously in both directions, resulting in a permanent collision. In
older forms of Ethernet, such a permanent collision, preclude the system from working.
Yet in 1000BASE-TX, sophisticated circuitry can accommodate this permanent colli-
sion, resulting from an attempt to get as much data as possible over UTP.
Media Access Control
Media Access Control (MAC) refers to protocols that determine which computer on a
shared-medium environment (collision domain) is allowed to transmit the data. MAC,
with LLC, comprises the IEEE version of Layer 2. MAC and LLC are both sublayers of
Layer 2. Two broad categories of MAC exist:
■ Deterministic (taking turns)
■ Nondeterministic (first come, first served)
Token Ring and FDDI are deterministic, and Ethernet/802.3 is nondeterministic
(also called probabilistic).
1102.book Page 266 Tuesday, May 20, 2003 2:53 PM
Ethernet Operation 267
Deterministic MAC Protocols
Deterministic MAC protocols use a form of taking turns. Token passing is example
of the deterministic MAC protocol. Some Native American tribes used the custom of
passing a “talking stick” during gatherings. Whoever held the talking stick was allowed
to speak. When that person finished, he or she passed it to another person.
In this analogy, the shared medium is the air, the data is the words of the speaker, and

the protocol is possession of the talking stick. The stick might even be called a token.
This situation is similar to a data link protocol called a Token Ring. In a Token Ring
network, individual hosts are arranged in a ring, as shown in Figure 5-9. A special data
token circulates around the ring. When a host wants to transmit, it seizes the token,
transmits the data for a limited time, and then places the token back in the ring, where
it can be passed along, or seized, by another host.
Figure 5-9 A Token Ring Network
Nondeterministic MAC Protocols
Nondeterministic MAC protocols use a first-come, first-served (FCFS) approach.
CSMA/CD is an example of a nondeterministic MAC protocol.
To use this shared-medium technology, Ethernet allows the networking devices to arbi-
trate for the right to transmit. Stations on a CSMA/CD network listen for quiet, at which
time it’s okay to transmit. However, if two stations transmit at the same time, a collision
occurs and neither station’s transmission succeeds. All other stations on the network
also hear the collision and wait for silence. The transmitting stations, in turn, each wait a
random period of time (a backoff period) before retransmitting, thus minimizing the
probability of a second collision.
Token
Ring
1102.book Page 267 Tuesday, May 20, 2003 2:53 PM
268 Chapter 5: Ethernet Fundamentals
Three Specific Topological Implementations and Their MACs
Three well-known Layer 2 technologies are Token Ring, FDDI, and Ethernet. Of these,
Ethernet is by far the most common; however, they all serve to illustrate a different
approach to LAN requirements. All three specify Layer 2 elements (for example, LLC,
naming, framing, and MAC), as well as Layer 1 signaling components and media
issues. The specific technologies for each are as follows:
■ Ethernet—Logical bus topology (information flow on a linear bus) and physical
star or extended star (wired as a star)
■ Token Ring—Logical ring topology (in other words, information flow controlled

in a ring) and a physical star topology (in other words, wired as a star)
■ FDDI—Logical ring topology (information flow controlled in a ring) and physical
dual-ring topology (wired as a dual-ring)
Ethernet MAC
Ethernet is a shared-media broadcast technology. The access method CSMA/CD used
in Ethernet performs three functions:
■ Transmitting and receiving data packets
■ Decoding data packets and checking them for valid addresses before passing
them to the upper layers of the OSI model
■ Detecting errors within data packets or on the network
In the CSMA/CD access method, networking devices with data to transmit over the
networking media work in a listen-before-transmit mode (CS = carrier sense). With
shared Ethernet, this means that when a device wants to send data, it first must check
to see whether the networking medium is busy. The device must check whether there
are any signals on the networking media. After the device determines that the network-
ing media is not busy, the device begins to transmit its data. While transmitting its data
in the form of signals, the device also listens, to ensure that no other stations are trans-
mitting data to the networking medium at the same time. If two stations send data at
the same time, a collision occurs, as shown in the upper half of Figure 5-10. When it
completes transmitting its data, the device returns to listening mode. With traditional
shared Ethernet, only one device can transmit at a time. This is not true with switched
Ethernet, which is covered in Chapter 6.
1102.book Page 268 Tuesday, May 20, 2003 2:53 PM