By Christina Morales,2014-07-12 13:16
12 views 0

    ? is the oldest layer 2 protocol

    designed by IBM in 1975 to carry

    traffic (see later). SDLC is the precursor to HDLC.

    ? was created out of SDLC by the

    ISO in 1979 (see later).

    ? was developed by IBM to

    enable SNA traffic over X.25.

    ? is one of the three layers

    making up X.25 and was a development by ITU-T, of LAP which itself

    came from HDLC. LAPB operates like HDLC, however it is restricted

    to using the ABM transfer mode i.e. point-to-point links (see


    ? provides HDLC services for

    V.42 modems.

    ? is used on the ISDN D

    channel for packet delivery.

    ? is used on the ISDN Terminal Adapters to allow multiple users

    across a single link.

    ? was developed by the IEEE 802.2, and is

    used to provide HDLC style services on a LAN (see below).

    ? , some operations are similar to LAPD and V.120.

    ? is a derivation of HDLC and encapsulates Protocol Data Units

    for transport across point-to-point links.

Although a subset of HDLC, SDLC was developed by IBM before HDLC, and was

    the first link layer protocol based on synchronous, bit-oriented

    operation. IBM defined SDLC for managing synchronous serially transmitted

    bits over a data link, and these links can be full/half-duplex, switched

or unswitched, point-to-point, point to multipoint or even looped. SDLC

    is designed for carrying SNA traffic.

In SDLC, a link station is a logical connection between adjacent nodes.

    Only one is allowed on an SDLC line. A device can

    be set up as a Primary or a Secondary link station. A device configured

    as a Primary link station can communicate with both PU 2.0 nodes and PU

    2.1 nodes (APPN) and controls the secondary devices. If the device is set

    up as a secondary link station then it acts as a PU 2.0 device and can

    communicate with Front End Processors (FEP), but only communicates with

    the primary device when the primary allows it, i.e. the primary sets up

    and tears down the connections and controls the secondaries.

In APPN configurations the device can support negotiable link stations

    where XID frames are exchanged to decide which station is to be secondary

    and which is to be primary.

A primary station issues commands, controls the link and initiates

    error-recovery. A device set up as a secondary station can communicate

    to a FEP, exist with other secondary devices on an SDLC link and exist

    as a secondary PU 2.0 device.

SDLC supports line speeds up to 64Kb/s e.g. V.24 (RS-232) at 19.2Kb/s,

    V.35 (up to 64Kb/s) and X.21.

The following diagram shows the frame format for SDLC, almost identical

    to HDLC.

? - Begins and ends the error checking procedure with which

    is in binary. ? - This is only the secondary address since all communication

    occurs via the single primary device. The address can be an

    individual, group or broadcast address.

    ? - this identifies the frame's function and can be one of

    the following:

    o - contains the which

    is the number of the next frame to be sent, and the

     which is the number of the next frame expected

    to be received. The is also a which

    performs error checking.

    o - this can report on status, ask for and stop

    transmission and acknowledge frames.

    o - this does not have sequence numbers (hence

    'unnumbered'), it can be used to start up secondaries and can

    sometimes have an Information field.

    ? - can contain or


    ? - this check is carried out on the sending

    AND receiving of the frame.

    In a poll, the address field identifies the station being polled, in a

    response, the address field contains the station transmitting, so this

    field effectively is the secondary station's address. The control field .

HDLC is the protocol which is now considered an umbrella under which many

    Wide Area protocols sit. ITU-T developed HDLC in 1979, and within HDLC

    there are three types of stations defined:

    ? - this completely controls all data link operations

    issuing commands from secondary stations and has the ability to hold

    separate sessions with different stations.

    ? - this can only send responses to one primary

    station. Secondary stations only talk to each other via a Primary


    ? - this can transmit and receive commands and

    responses from one other station.

    Configuring a channel for use by a station can be done in one of three


    ? - this configuration allows one primary station to talk

    to a number of secondary stations over half-duplex, full-duplex,

    switched, unswitched, point-to-point or multipoint paths.

? - where commands and responses are multiplexed over one

    physical channel when two stations with primary and secondary parts

    have a point-to-point link joining them.

    ? - where two combined stations communicate over a

    point-to-point link which can be full/half-duplex or


    When transferring data, stations are in one of three modes:

    ? where the secondary station needs

    permission from the primary station before it can transmit data.

    Mainly used on multi-point lines.

    ? where the secondary station can

    send data without receiving permission from the primary station.

    This is hardly ever used.

    ? where either station can initiate

    transmission without permission from the other. This is the most

    common mode used on point-to-point links.

    The following diagram details the HDLC frame format:

The HDLC frame begins and ends the error checking procedure with which

    is in binary.

There are three types of HDLC frame types defined by the control field:

? are used for the data transfer between stations.

    The send sequence, or next send N(S), and the receive sequence, or

    next receive N(R), hold the frame sequence numbers. The

    bit is called Poll when used by the primary station to obtain a

    response from a secondary station, and Final when used by the

    secondary station to indicate a response or the end of transmission.

    ? are used to acknowledge frames, request for

    retransmissions or to ask for suspension of transmission. The

    Supervisory code denotes the type of supervisory frame being sent.

    ? are used for link initialisation or link

    disconnection. The Unnumbered bits indicate the type of Unnumbered

    frame being used.

LLC is a subset of and uses the

     subclass of HDLC. It sits in the Data Link layer between the MAC layer and the layer 3 protocols and forms an

    important part of the 802.2 specification.

There are three Classes of LLC:

    1. - , this service sends and receives

     without the need for acknowledgement.

    It supports point-to-point, multipoint and broadcast communication

    and is suitable for higher level protocols that do all the

    sequencing, addressing, routing and recovery. These include IPX,

    TCP/IP, Vines, XNS, AppleTalk etc. LLC1 is less intense on network

    resources than LLC2 or LLC3 as there is less requirement for the

    data-integrity measures included within the following two methods.

    The responsibility for data flow and integrity rests with the layer

    3 protocols. LLC1 operates operations which means that

     are exchanged without connections, there

    is no sequencing, acknowledgement or error-checking of PDUs.

    2. - , this is a

    connection-oriented service that provides a point-to-point link

    between . Although LLC2 responds

    to the higher layer protocol with respect to opening and closing

    connections, LLC2 is responsible for flow control, sequencing the

    frames and error recovery. LLC2 can operate or

    operations, where a connection has to be established,

    acknowledgement, sequencing and error checking all take place in

    the LLC sublayer. LLC2 is generally required in environments that

    run protocols such as NetBIOS and SNA.

    3. - , this is a

    'connectionless-acknowledged' implementation that is rarely used.

    The following diagram shows the structure of the LLC PDU.

This sits at the start of the data part of the Ethernet frame (see the

    IPX/SPX document).

The following table lists the LLC command PDUs:

    Type 1 Unnumbered Unnumbered 03 (CL) Information (UI) (U)

    Exchange Exchange Unnumbered Identification Identification AF, BF (U) (XI) (XI)

    Unnumbered Test (TEST) Test (TEST) E, F3 (U) Type 2 Information 00 00 to Information (I) Information (I) (CO) (I) FE FF

    Receiver Ready Receiver Ready Supervisory 01 00 to (RR) (RR) (S) 01 FF

    Receiver Not Ready Receiver Not Supervisory 05 00 to (RNR) Ready (RNR) (S) 05 FF

    Supervisory 09 00 to Reject (REJ) Reject (REJ) (S) 09 FF

    6F, 7F Set Asynchronous Unnumbered Unnumbered (SABME) Balance Mode Acknowledgement (U) and 63, Extended (SABME) (UA) 73 (UA)

    43, 53

    Disconnected Unnumbered (DISC) Disconnect (DISC) Mode (DM) (U) and 0F,

    1F (DM)

    Frame Reject Unnumbered 87, 97 (FRMR) (U)

    Ack Ack Type 3 Unnumbered Connectionless, Connectionless, 67, F7 (AC) (U) seq 0 (AC0) seq 0 (AC0)

    Ack Ack Unnumbered Connectionless, Connectionless, E7, F7 (U) seq 0 (AC1) seq 0 (AC1)

The LLC SAP's job is to sort the up and coming MAC frames and direct them

    to the appropriate application or upper layer protocol. NetBIOS uses the

    SAP address of , IP uses whilst DLSw uses and Network Management

    uses . The , address does not identify any SAP to any protocol.

A more complete list follows:

    00 Null SAP

    04 SNA

    05 SNA

    06 TCP

    08 SNA

    0C SNA

    42 Spanning Tree

    7F ISO 802.2

    80 XNS


    E0 IPX

    F0 NetBIOS

    F8 RPL

    FC RPL

    FE OSI

    FF Global SAP

The complete list can be found at

The Control field identifies sequencing and/or commands/responses. There

    are three types:

    ? The carries out a numbered information

    transfer in Type 2 operations.

    ? The carries out acknowledgement, retransmit

    requests and suspends of I PDUs in Type 2 operations.

    ? The carries out some control functions and

    unsequenced information transfer in Type 1 operations.

    There is the special case of where the

    DSAP and SSAP are set to and the Control field set to . Then,

    immediately after the control field there are three bytes set aside for

    the and then two bytes for the

     (see the IPX/SPX document).

The SAP is generally used for the 802.x compliant protocols, however, the

    idea of SNAP is to allow non-IEEE compliant protocols to become

    pseudo-compliant without a major re-write of code for network drivers.

    An OUI of 00-00-00 indicates that the frame is an Ethernet frame rather

    than an 802.3 frame that is not assigned to any particular vendor. This

    useful for when the frame crosses different media. On coming out the other

    side it will be rebuilt as an Ethernet frame rather than something else.

Report this document

For any questions or suggestions please email