ATM (Asynchronous Transfer Mode) Tutorial-Network Architecture,cell,addressing,interfaces
This page of tutorials section covers ATM (Asynchronous Transfer Mode) tutorial. It covers ATM cell, ATM addressing formats, ATM network architecture and ATM network interfaces etc. The Asynchronous Transfer Mode tutorial provides link to ATM protocol layers, ATM services etc.
Refer Following links to subtopics on this ATM tutorial:
ATM Network Architecture & Interfaces ATM Protocol Stack ATM services ATM Switch ATM Call Flow ATM vs STM ATM vs TDM ATM versus Frame Relay ATM AAL Layer Types
In the year 1980, ITU-T has initiated efforts to develop a network which can transport voice, video and data simultaneously from source to destination. As a result B-ISDN (Broadband Integrated Services Digital Network) has been developed. The B-ISDN combines telephony network, data network and cable TV network in a single network. ATM as well as SONET/SDH are behind the success of B-ISDN.
The same is shown in the figure-1. Unlike Synchronous Transfer Mode where in each source will get periodic assignment of bandwidth similar to TDM, Source in ATM packetizes data into smaller fixed size cells and cells are sent only if there is data to be transmitted. In ATM each of the cells will have its own header as described later.
Following are the silent features of ATM (Asynchronous Transfer Mode):
• ATM uses Virtual Circuit Packet Switching. It reserves capacity for the virtual circuit.
• Packets are known as cells in ATM. The smaller packets are good for voice as well as video transmissions. One ATM cell consists of header (5 bytes) and data payload (48 bytes).
• It allows multiple logical connections to be multiplexed over single physical interface.
• It is a connection oriented technology.
• As mentioned it provides dynamic allocation of bandwidth for efficient traffic management.
• Supports minimum flow control and error control capabilities.
• Supports LAN, CAN and WAN
• ATM can be deployed in private, public or hybrid networks.
• In ATM (Asynchronous Transfer Mode) logical connections are known as virtual channel connections or VCCs.
The VCC is similar to virtual circuit (VC) in packet switched network.
It is unit of switching used in ATM network.
• Initially VCC is established between end users in the network.
• After the VCC connection is established variable rate and fixed size cells are exchanged.
• Here VPC i.e. Virtual Path Connection is combination of VCCs having same end points or destinations. Hence all the ATM cells travelling through VCCs in one common VPC are switched together.
• Virtual path is identified by VPI and Virtual Channel is identified by VCI as mentioned below.
➨VPI (Virtual Path Identifier) identifies virtual path (8/12 bits in size).
➨VCI (Virtual Channel Identifier) identifies virtual channel in a virtual path (16 bits).
ATM Network Architecture and network interfaces
• The figure-2 depicts ATM (Asynchronous Transfer Mode) network architecture with interfaces.
There are two types of Virtual connections supported by ATM network viz.
PVCs and SVCs.
• PVCs are referred as Permanent Virtual Connections. They are similar to leased lines used between users. The PVCs are set up by the operator.
• SVCs are referred as Switched Virtual Connections. They are set up and teared down based on end user demand.
• The ATM network interfaces include UNI (User-Network Interface) , NNI (Network-Network Interface) and B-ICI (Broadband Intercarrier Interface). User interacts with network using UNI interface to establish SVC connection. The ATM switches interact using NNI interface in order to exchange information. ATM switches belonging to another public networks will communicate using B-ICI interface.
ATM Cell-UNI cell vs NNI cell
As mentioned earlier ATM cell is composed of header and payload part. Header is of size equal to 5 bytes or octets and payload which carrier information from upper layers is of size equal to 48 bytes. Header of ATM (Asynchronous Transfer Mode) cell varies in UNI interface and NNI interface. The same have been shown in the figure below. Let us understand difference between UNI cell header and NNI cell header. As shown, GFC field is used only in UNI cells. At NNI, GFC byte is used to incorporate additional VPI.
|UNI Cell Header field||Size||Description|
|GFC||4 bits||Generic Flow Control|
|VPI||(8 bits for UNI or 12 bits for NNI)||Virtual Path Identifier, It is a routing field for the network.|
|VCI||16 bits||Virtual Channel Identifier, It is used for routing to and from end user. It acts as SAP(Service Access Point).|
|PT||3 bits||Payload Type|
|CLP||1 bit||Cell Loss Priority , value of 0 indicates that cell should not be discarded. Value of 1 indicates that cell can be discarded.|
|HEC||8 bits||Header Error Control, It is used to correct single bit errors in the header. It is also used to detect double bit errors.|
|NNI Cell Header field||Size||Description|
|VPI||8/12 bits||Virtual Path Identifier|
|VCI||16 bits||Virtual Channel Identifier|
|PT||3 bits||Payload Type|
|CLP||1 bit||Cell Loss Priority|
|HEC||8 bits||Header Error Control|
ATM Addressing Formats-DCC, ICD, E.164
All the ATM (Asynchronous Transfer Mode) addresses are of 20 byte in size.
Source and destination addresses are used while establishing a connection.
There are three types of formats used for ATM address viz. NSAP (Network Service Access Point), DCC and ICD.
• ATM end points use NSAP format. ISDN telephone numbers use the format E.164.
• DCC (Data Country Code) format is used for public networks.
• ICD international Code Designator) format is used for private networks.
Following table describes fields used in ATM addresses.
|AFI||1 byte||Authority and Format Identifier, tells which addressing scheme is used.|
|IDI||2-8 bytes||Initial Domain Identifier, Identifies a domain within scope of addressing authority.|
|HO-DSP||4-10 bytes||high Order bits of domain specific part, same as network prefix of IP address|
|ESI||6 bytes||End System Identifier, Same as host number of IP address|
|SEL||1 byte||Selector, for endsystem use only|
ATM Protocol Stack
ATM protocol stack consists of three main layers viz. AAL Layer, ATM layer and physical layer.
Above AAL layer upper layers reside. Following are the functions of each.
➨AAL layer does encapsulation of user level data. AAL layer breaks the upper layer data at the sender side into small size ATM cells and reassembles the cells at the receiver side.
➨ATM layer take care of transport of 53 byte ATM cells created by AAL layer across the ATM network. It does multiplexing of multiple logical channels on one single physical channel. It provides different services for ATM cells.
➨ATM physical layer does encoding (at transmit end) and decoding (at receive end) of bits based on physical medium used for transportation of ATM cells.
Refer ATM Protocol Stack➤ which covers Asynchronous Transfer Mode Protocol layers, their sublayers and functions.
ATM Services-CBR, rt-VBR, nrt-VBR, ABR, UBR, GFR
The ATMM layer provides following different kinds of ATM services for ATM cells as per requirement in different applications. The table mentions the same.
|ATM Service Type||Description|
|CBR-Constant Bit Rate||• guarantees a fixed capacity, similar to circuit switching
• guarantees a maximum delay for cells
|VBR-Variable Bit Rate||
• Supports both realtime and nonrealtime applications
• guarantees an average throughput and maximum delay
|ABR-Available Bit Rate||• guarantees "fairness" with respect to other traffic|
|UBR-Unspecified Bit Rate||• service is on a "best effort" basis|
|GFR-Guarantees Frame Rate||• Throughput guarantee for multiple cell frames|
Refer ATM services➤ for more information.