Biblio5

534 ASYNCHRONOUS TRANSFER MODE

Figure 18.1 An ATM link simultaneously carries a mix of voice, data, and image information. (From Ref. 1, courtesy of Hewlett-Packard Co.)

not the last, are identified with the COM (continuation of message), and the last piece or slot is identified with an EOM (end of message). There is also an MID (message identifier) appended to all pieces of a common message. This is done so that there is no confusion as to which transaction a cell belongs.

18.2 INTRODUCTION TO ATM

ATM is an outgrowth of the several data transmission format systems mentioned in the preceding section, although some may argue this point. Whereas the formats described ostensibly were to satisfy the needs of the data world, ATM (according to some) provides an optimum format or protocol family for data, voice, and image communications, where cells of each medium can be intermixed through the network, as illustrated in Figure 18.1.3 It would really seem to be more of a compromise from our perspective. Typically, these ATM cells can be transported on SONET, SDH, E1/ DS1, and on other digital formats. Cells can also be transported contiguously without an underlying digital network format.

Philosophically, voice and data are worlds apart regarding time sensitivity. Voice cannot wait for long processing and ARQ delays. Most types of data can. So ATM must distinguish the type of service such as constant bit rate (CBR) and variable bit rate (VBR) services. Voice service is typical of constant bit rate or CBR service.

Signaling is another area of major philosophical difference. In data communications, “signaling” is carried out within the header of a data frame (or packet). As a minimum the signaling will have the destination address, and quite often the source address as well. And this signaling information will be repeated over and over again on a long data file that is heavily segmented. On a voice circuit, a connectivity is set up and the destination address, and possibly the source address, is sent just once during call setup. There is also some form of circuit supervision to keep the circuit operational throughout the duration of a telephone call. ATM is a compromise, stealing a little from each of these separate worlds.

Like voice telephony, ATM is fundamentally a connection-oriented telecommunication system. Here we mean that a connection must be established between two stations before data can be transferred between them. An ATM connection specifies the transmission path, allowing ATM cells to self-route through an ATM network. Being

3DQDB can also transport voice in its PA (pre-arbitrated) segments.

connection-oriented also allows ATM to specify a guaranteed quality of service (QoS) for each connection.

By contrast, most LAN protocols are connectionless. This means that LAN nodes simply transmit traffic when they need to, without first establishing a specific connection or route with the destination node.

In that ATM uses a connection-oriented protocol, cells are allocated only when the originating end-user requests a connection. They are allocated from an idle-cell pool. This allows ATM to efficiently support a network’s aggregate demand by allocating cell capacity on demand based on immediate user need. Indeed it is this concept that lies at the heart of the word asynchronous (as in asynchronous transfer mode). An analogy would help. Let’s say that New York City is connected to Washington, DC, by a pair of railroad tracks for passenger trains headed south and another pair of tracks for passenger trains headed north. On these two pair of tracks we’d like to accommodate everybody we can when they’d like to ride. The optimum for reaching this goal is to have a continuous train of coupled passenger cars. As the train enters Union Station, it discharges its passengers and connects around directly for the northward run to Pennsylvania Station and loading people bound for New York. Passenger cars are identical in size, and each has the same number of identical seats.

Of course, at 2 A.M. the train will have very few passengers and many empty seats. Probably from 7 A.M. to 9 A.M. the train will be full, no standees allowed, so we’ll have to hold potential riders in the waiting room. They’ll ride later; those few who try to be standees will be bumped. Others might seek alternate transportation to Washington, DC.

Here we see that the railroad tracks are the transmission medium. Each passenger car is a SONET/ SDH frame. The seats in each car are our ATM cells. Each seat can handle a person no bigger than 53 units. Because of critical weight distribution, if a person is not 53 units in size/ weight, we’ll stick some bricks in the seat to bring the size/ weight to 53 units exactly. Those bricks are removed at the destination. All kinds of people ride the train because America is culturally diverse, analogous to the fact that ATM handles all forms of traffic. The empty seats represent idle or unassigned cells. The header information is analogous to the passengers’ tickets. Keep in mind that the train can only fill to its maximum capacity of seats. We can imagine the SONET/ SDH frame as being full of cells in the payload, some cells busy and some idle/ unassigned. At the peak traffic period, all cells will be busy, and some traffic (passengers) may have to be turned away.

We can go even further with this analogy. Both Washington, DC, and New York City attract large groups of tourists, and other groups travel to business meetings or conventions. A tour group has a chief tour guide in the lead seat (cell) and an assistant guide in the last seat (cell). There may be so many in the group that they extend into a second car or may just intermingle with other passengers on the train. The tour guide and assistant tour guide keep an exact count of people on the tour. The lead guide wears a badge that says BOM, all tour members wear badges that say COM, and the assistant tour guide wears a badge that says EOM. Each group has a unique MID (message ID). We also see that service is connection-oriented (Washington, DC, to New York City).

Asynchronous means that we can keep filling seats on the train until we reach its maximum capacity. If we look up the word, it means nonperiodic, whereas the familiar E1/ DS1 is periodic (i.e., synchronous). One point that seems to get lost in the literature is that the train has a maximum capacity. It is common to read that ATM provides “bandwidth” on demand. The underlying transmission medium has a fixed bandwidth. We believe the statement is supposed to mean “cell rate capacity” on demand. This is true, of course, until we reach the maximum capacity. For example, if the underlying

536 ASYNCHRONOUS TRANSFER MODE

Figure 18.2 ATM reference model and user-network interface configuration. (Sources: Refs. 2, 3, and 6.)

transmission format is SONET STS-1, it can provide 84 × 9 octets × 8000/ (53) per second or about 114 thousand cells per second of payload (see Chapter 17).

18.3 USER–NETWORK INTERFACE (UNI) AND ARCHITECTURE

ATM is the underlying packet technology of broadband ISDN (B-ISDN). At times in this section, we will use the terms ATM and B-ISDN interchangeably. Figures 18.2 and 18.3 interrelate the two. Figure 18.2 relates the B-ISDN access reference configuration with the ATM user–network interface (UNI). Note the similarities of this figure with Figure 12.15, the reference model for ISDN. The only difference is that the block nomenclature has a “B” placed in front to indicate broadband. Figure 18.3 is the traditional CCITT Rec. I.121 (Ref. 2) B-ISDN protocol reference model, showing the extra layer necessary for the several services.

Returning to Figure 18.2, we see that there is an upper part and a lower part. The lower part shows the UNI boundaries. The upper part is the B-ISDN reference configuration with four interface points. The interfaces at the reference points UB, TB, and SB are standardized. These interfaces support all B-ISDN services.

There is only one interface per B-NT1 at the UB and one at the TB reference points. The physical media is point-to-point (in each case), in the sense that there is only one receiver in front of one transmitter.

One or more interfaces per NT2 are present at the SB reference point. The interface at the SB reference point is point-to-point at the physical layer, in the sense that there is only one receiver in front of one transmitter, and may be point-to-point at other layers.

Consider now the functional groupings in Figure 18.2. B-NT1 includes functions broadly equivalent to OSI layer 1, the physical layer. These functions include:

Figure 18.3 B-ISDN reference model. (From Refs. 4, 6, and 7.)

•Line transmission termination;

•Interface handling at TB and UB; and

•OAM functions.4

The B-NT2 functional group includes functions broadly equivalent to OSI layer 1 and higher OSI layers. The B-NT2 may be concentrated or distributed. In a particular access arrangement, the B-NT2 functions may consist of physical connections. Examples of B-NT2 functions are:

•Adaptation functions for different media and topologies;

•Cell delineation;

•Concentration; buffering;

•Multiplexing and demultiplexing;

•OAM functions;

•Resource allocation; and

•Signaling protocol handling.

The functional group B-TE (TE stands for terminal equipment) also includes functions of OSI layer 1 and higher OSI layers. Some of these functions are:

• User/ user and user/ machine dialog and protocol;

• Protocol handling for signaling;

4OAM stands for operations, administration, and maintenance.

540 ASYNCHRONOUS TRANSFER MODE

side of Figure 18.5 shows the structure of a UNI header, whereas the right-hand side illustrates the NNI header. The only difference is the presence of the GFC (generic flow control) field in the UNI header. The following paragraphs define each header field. By removing the GFC field, the NNI has four additional bits for addressing.

18.4.1.1 GFC–Generic Flow Control. The GFC field contains 4 bits. When the GFC function is not used, the value of this field is 0000. This field has local significance only and can be used to provide standardized local flow control functions on the customer side. In fact, the value encoded in the GFC is not carried end-to-end and will be overwritten by ATM switches (i.e., the NNI interface).

Two modes of operation have been defined for operation of the GFC field. These are uncontrolled access and controlled access. The uncontrolled access mode of operation is used in the early ATM environment. This mode has no impact on the traffic which a host generates. Each host transmits the GFC field set to all zeros (0000). In order to avoid unwanted interactions between this mode and the controlled access mode, where hosts are expected to modify their transmissions according to the activity of the GFC field, it is required that all CPE (customer premise equipment) and public network equipment monitor the GFC field to ensure that attached equipment is operating in uncontrolled mode. A count of the number of nonzero GFC fields should be measured for nonoverlapping intervals of 30,000 ± 10,000 cell times. If ten or more nonzero values are received within this interval, an error is indicated to layer management (Ref. 3).

18.4.1.2 Routing Field (VPI/ VCI). Twenty-four bits are available for routing a cell. There are 8 bits for virtual path identifier (VPI) and 16 bits for virtual channel identifier (VCI). Preassigned combinations of VPI and VCI values are given in Table 18.1. Other preassigned values of VPI and VCI are for further study, according to the ITU-T Organization. The VCI value of zero is not available for user virtual channel identification. The bits within the VPI and VCI fields are used for routing and are allocated using the following rules:

•The allocated bits of the VPI field are contiguous.

•The allocated bits of the VPI field are the least significant bits of the VPI field, beginning at bit 5 of octet 2.

•The allocated bits of the VCI field are contiguous.

•The allocated bits of the VCI field are the least significant bits of the VCI field, beginning at bit 5 of octet 4.

18.4.1.3 Payload Type (PT) Field. Three bits are available for PT identification. Table 18.2 gives the payload type identifier (PTI) coding. The principal purpose of the PTI is to discriminate between user cells (i.e., cells carrying information) and nonuser cells. The first four code groups (000 –011) are used to indicate user cells. Within these four, 2 and 3 (010 and 011) are used to indicate that congestion has been experienced. The fifth and sixth code groups (100 and 101) are used for virtual channel connection (VCC) level management functions.

Any congested network element, upon receiving a user data cell, may modify the PTI as follows: Cells received with PTI c 000 or PTI c 010 are transmitted with PTI c 010. Cells received with PTI c 001 or PTI c 011 are transmitted with PTI c 011. Noncongested network elements should not change the PTI.

Source: ITU-T Rec. I.432, Table 4/ I.432, p. 19 (Ref. 4).

18.5 CELL DELINEATION AND SCRAMBLING

Cell delineation allows identification of the cell boundaries. The cell HEC field achieves cell delineation. Keep in mind that the ATM signal must be self-supporting, in that it has to be transparently transported on every network interface without any constraints from the transmission systems used. Scrambling is used to improve security and robustness of the HEC cell delineation mechanism discussed in the next paragraph. In addition, it helps the randomizing of data in the information field for possible improvement in transmission performance.

Cell delineation is performed by using correlation between the header bits to be protected (the first four octets in the header) and the HEC octet. This octet is produced at the originating end using a generating polynomial covering those first four octets of the cell. The generating polynomial is X8 + X 2 + X + 1. There must be correlation at the receiving end between those first four octets and the HEC octet, which we can call a remainder. This is only true, of course, if there is no error in the header. When there is an error, correlation cannot be achieved, and the processor just goes to the next cell.

18.6 ATM LAYERING AND B-ISDN

The B-ISDN reference model was given in Figure 18.3, and its several planes were described. This section provides brief descriptions of the ATM layers and sublayers. Figure 18.7 illustrates B-ISDN/ ATM layering and sublayering of the protocol reference model. It identifies functions of the physical layer, the ATM layer, the AAL (ATM adaptation layer), and related sublayers.

18.6.1 Physical Layer

The physical layer consists of two sublayers. The physical medium (PM) sublayer includes only physical medium-dependent functions. The transmission convergence (TC) sublayer performs all functions required to transform a flow of cells into a flow of data units (i.e., bits), which can be transmitted and received over a physical medium. The service data unit (SDU) crossing the boundary between the ATM layer and the physical layer is a flow of valid cells. The ATM layer is unique (independent of the underlying physical layer). The data flow inserted in the transmission system payload is physical medium-independent and self-supported. The physical layer merges the ATM cell flow with the appropriate information for cell delineation, according to the cell delineation mechanism previously described, and carries the operations administration and maintenance (OAM) information relating to this cell flow.

The PM sublayer provides bit transmission capability including bit transfer and bit alignment, as well as line coding and electrical–optical transformation. Of course, the principal function is the generation and reception of waveforms suitable for the medium, the insertion and extraction of bit timing information, and line coding where required.