In this section, we outline several commonly used data link layer protocols, which are summarized in Figure 4.7. Here we focus on message delineation, which indicates where a message starts and stops, and the various parts or fields within the message. For example, you must clearly indicate which part of a message or packet of data is the error-control portion; otherwise, the receiver cannot use it properly to determine if an error has occurred. The data link layer performs this function by adding a PDU to the packet it receives from the network layer. This PDU is called frame.
Asynchronous Transmission
Asynchronous transmission often is referred to as start-stop transmission because the transmitting computer can transmit a character whenever it is convenient, and the receiving computer will accept that character.
Figure 4.6 Hamming code for forward error correction
|
Protocol |
Size |
Error Detection |
Retransmission |
Media Access |
|
Asynchronous transmission |
1 |
Parity |
Continuous ARQ |
Full Duplex |
|
Synchronous protocols |
||||
|
SDLC |
* |
16-bit CRC |
Continuous ARQ |
Controlled Access |
|
HDLC |
* |
16-bit CRC |
Continuous ARQ |
Controlled Access |
|
Ethernet |
* |
32-bit CRC |
Stop-and-wait ARQ |
Contention |
|
PPP |
* |
16-bit CRC |
Continuous ARQ |
Full Duplex |
Varies depending on the message length. ARQ = Automatic Repeat reQuest; CRC = cyclical redundancy check; HDLC = high-level data link control; PPP = Point-to-Point Protocol; SDLC = synchronous data link control.
Figure 4.7 Protocol summary
It is typically used on point-to-point full-duplex circuits (i.e., circuits that have only two computers on them), so media access control is not a concern. If you use VT100 protocol, or connect to a UNIX or Linux computer using Telnet, chances are you are using asynchronous transmission.
With asynchronous transmission, each character is transmitted independently of all other characters. To separate the characters and synchronize transmission, a start bit and a stop bit are put on the front and back of each individual character. For example, if we are using 7-bit ASCII with even parity, the total transmission is 10 bits for each character (1 start bit, 7 bits for the letter, 1 parity bit, 1 stop bit).
Figure 4.8 Asynchronous transmission. ASCII = United States of America Standard Code for Information Interchange
The start bit and stop bit are the opposite of each other. Typically, the start bit is a 0 and the stop bit is a 1. There is no fixed distance between characters because the terminal transmits the character as soon as it is typed, which varies with the speed of the typist. The recognition of the start and stop of each message (called synchronization) takes place for each individual character because the start bit is a signal that tells the receiver to start sampling the incoming bits of a character so the data bits can be interpreted into their proper character structure. A stop bit informs the receiver that the character has been received and resets it for recognition of the next start bit.
When the sender is waiting for the user to type the next character, no data is sent; the communication circuit is idle. This idle time really is artificial—some signal always must be sent down the circuit. For example, suppose we are using a unipolar digital signaling technique where +5 volts indicates a 1 and 0 volts indicates a 0.Even if we send 0 volts, we are still sending a signal, a 0 in this case. Asynchronous transmission defines the idle signal (the signal that is sent down the circuit when no data are being transmitted) as the same as the stop bit. When the sender finishes transmitting a letter and is waiting for more data to send, it sends a continuous series of stop bits. Figure 4.8 shows an example of asynchronous transmission.
Some older protocols have two stop bits instead of the traditional single stop bit. The use of both a start bit and a stop bit is changing; some protocols have eliminated the stop bit altogether.
Synchronous Transmission
With synchronous transmission, all the letters or data in one group of data is transmitted at one time as a block of data. This block of data is called a frame. For example, a terminal or microcomputer will save all the keystrokes typed by the user and transmit them only when the user presses a special "transmit" key. In this case, the start and end of the entire frame must be marked, not the start and end of each letter. Synchronous transmission is often used on both point-to-point and multipoint circuits. For multipoint circuits, each packet must include a destination address and a source address, and media access control is important.
The start and end of each frame (synchronization) sometimes is established by adding synchronization characters (SYN) to the start of the frame. Depending on the protocol, there may be anywhere from one to eight SYN characters. After the SYN characters, the transmitting computer sends a long stream of data that may contain thousands of bits. Knowing what code is being used, the receiving computer counts off the appropriate number of bits for the first character, assumes this is the first character, and passes it to the computer. It then counts off the bits for the second character, and so on.
Figure 4.9 SDLC (synchronous data link control) frame layout
In summary, asynchronous data transmission means each character is transmitted as a totally independent entity with its own start and stop bits to inform the receiving computer that the character is beginning and ending. Synchronous transmission means whole blocks of data are transmitted as frames after the sender and the receiver have been synchronized.
There are many protocols for synchronous transmission. We discuss four commonly used synchronous data link protocols.
Synchronous Data Link Control Synchronous data link control (SDLC) is a mainframe protocol developed by IBM in 1972 that is still in use today. It uses a controlled-access media access protocol. If you use a 3270 protocol, you’re using SDLC.
Figure 4.9 shows a typical SDLC frame. Each SDLC frame begins and ends with a special bit pattern (01111110), known as the flag. The address field identifies the destination. The length of the address field is usually 8 bits but can be set at 16 bits; all computers on the same network must use the same length. The control field identifies the kind of frame that is being transmitted, either information or supervisory. An information frame is used for the transfer and reception of messages, frame numbering of contiguous frames, and the like. A supervisory frame is used to transmit acknowledgments (ACKs and NAKs). The message field is of variable length and is the user’s message. The frame check sequence field is a 32-bit CRC code (some older versions use a 16-bit CRC).
High-Level Data Link Control High-level data link control (HDLC) is a formal standard developed by the ISO often used in WANs. HDLC is essentially the same as SDLC, except that the address and control fields can be longer. HDLC also has several additional benefits that are beyond the scope of this topic, such as a larger sliding window for continuous ARQ. It uses a controlled-access media access protocol. One variant, Link Access Protocol-Balanced (LAP-B), uses the same structure as HDLC but is a scaled-down version of HDLC (i.e., provides fewer of those benefits mentioned that are "beyond the scope of this topic"). A version of HDLC called Cisco HDLC (cHDLC) includes a network protocol field. cHDLC and HDLC have gradually replaced SDLC.
Ethernet Ethernet is a very popular LAN protocol, conceived by Bob Metcalfe in 1973 and developed jointly by Digital, Intel, and Xerox in the 1970s. Since then, Ethernet has been further refined and developed into a formal standard called IEEE 802.3ac. There are several versions of Ethernet in use today. Ethernet uses a contention media access protocol.
There are several standard versions of Ethernet. Figure 4.10a shows an Ethernet 803.3ac frame. The frame starts with a 7-byte preamble which is a repeating pattern of ones and zeros (10101010).
FIGURE 4.10a Ethernet 802.3ac frame layout
FIGURE 4.10b Ethernet II frame layout
This is followed by a start of frame delimiter, which marks the start of the frame. The destination address specifies the receiver, whereas the source address specifies the sender. The length indicates the length in 8-bit bytes of the message portion of the frame. The VLAN tag field is an optional four-byte address field used by virtual LANs (VLANs), which are discussed in next topic. The Ethernet frame uses this field only when VLANs are in use; otherwise the field is omitted, and the length field immediately follows the source address field. When the VLAN tag field is in use, the first two bytes are set to the number 24,832 (hexadecimal 81-00), which is obviously an impossible packet length. When Ethernet sees this length, it knows that the VLAN tag field is in use. When the length is some other value, it assumes that VLAN tags are not in use and that the length field immediately follows the source address field. The DSAP and SSAP are used to pass control information between the sender and receiver. These are often used to indicate the type of network layer protocol the packet contains.The control field is used to hold the frame sequence numbers and ACKs and NAKs used for error control, as well as to enable the data link layers of communicating computers to exchange other control information. The last 2 bits in the first byte are used to indicate the type of control information being passed and whether the control field is 1 or 2 bytes (e.g., if the last 2 bits of the control field are 11, then the control field is 1 byte in length). In most cases, the control field is 1-byte long. The maximum length of the message is about 1,500 bytes. The frame ends with a CRC-32 frame check sequence used for error detection.
Ethernet II is another commonly used version of Ethernet. Like SDLC, it uses a preamble and a flag to mark the start and end of the frame. It has the same source and destination address format as Ethernet 802.3ac. The type field is used to specify an ACK frame or the type of network layer packet the frame contains (e.g., IP). The data and frame check sequence fields are the same as Ethernet 802.3ac.
Newer versions of these two types of Ethernet permits jumbo frames with up to 9,000 bytes of user data in the message field. Some vendors are experimenting with super jumbo frames that can hold up to 64,000 bytes. Neither jumbo nor super jumbo frame sizes have been standardized, but jumbo frames are common for some types of Ethernet such as gigabit Ethernet.
Point-to-Point Protocol Point-to-Point Protocol (PPP) was developed in the early 1990s and is often used in WANs. It is designed to transfer data over a point-to-point circuit but provides an address so that it can be used on multipoint circuits. Figure 4.11 shows the basic layout of a PPP frame, which is very similar to an SDLC or HDLC frame.
A Day in the Life: Network Support Technician
When a help call arrives at the help desk, the help desk staff (first-level support) spends up to 10 minutes attempting to solve the problem. If they can’t, then the problem is passed to the second-level support, the network support technician.
A typical day in the life of a network support technician starts by working on computers from the day before. Troubleshooting usually begins with a series of diagnostic tests to eliminate hardware problems. The next step, for a laptop, is to remove the hard disk and replace it with a hard disk containing a correct standard image. If the computer passes those tests then the problem is usually software. Then the fun begins.
Once a computer has been fixed it is important to document all the hardware and/or software changes to help track problem computers or problem software. Sometimes a problem is new but relatively straightforward to correct once it has been diagnosed. In this case, the technician will change the standard support process followed by the technicians working at the help desk to catch the problem before it is escalated to the network support technicians. In other cases, a new entry is made into the organization’s technical support knowledge base so that if another technician (or user) encounters the problem it is easier for him or her to diagnose and correct the problem. About 10% of the time the network technician is spent documenting solutions to problems.
Network support technicians also are the ones who manage new inventory and set up and configure new computers as they arrive from the manufacturer. They are also the ones responsible for deploying new software and standard desktop images across the network. Many companies also set aside standard times for routine training; in our case, every Friday, several hours is devoted to regular training.
Figure 4.11 PPP frame layout
The frame starts with a flag, and has a one-byte address (which is not used on point-to-point circuits). The control field is typically not used. The protocol field indicates what type of data packet the frame contains (e.g., an IP packet). The data field is variable in length and may be up to 1,500 bytes. The frame check sequence is usually a CRC-16, but can be a CRC-32. The frame ends with a flag.
