ATM Asynchronous Transfer Mode

- Fundamentals -

ATM overview

ATM (also known as cell relay) is a fast packet switching and multiplexing technology

ATM is a connection oriented, error-detecting protocol, but it does not offer error correction

ATM provides efficient support for transmission of bursty wideband services and offers an integrated solution to voice, data and video

ATM provides quality of service guarantee ad reliability

ATM uses statistical multiplexing (multiplexing many connection with variable rate characteristics) to take advantage of the bursty nature of the applications

The fundamental strategy behind ATM is to split the information into small fixed-sized units called cells which are easier to handle and provide more efficient switching and decrease overall bandwidth requirements

ATM cell

The user traffic is split and delivered in fixed length packets called ATM cells The size of the cell is 53 bytes, which is divided into a 5-byte header and 48-byte payload field The ATM cell is relayed by a label at the header:

Virtual Channel Identifier (VCI) Virtual path Identifier (VPI)

There are two formats of an ATM cell (depending on type of the interface)

ATM UNI (User-Network Interface) cell used for communication between ATM endpoints and ATM switches ATM NNI (Network-Node Interface) cell used for communication between ATM switches

Example of ATM interfaces in 3G network

UNI refers to the interface between terminal equipment and a network termination where access protocols apply The interface between a RNC and a WCDMA BTS is seen as a UNI interface NNI is the interface between two network nodes like a RNC and a MGW

Basic ATM cell format

There is a slight difference between the first byte of the UNI and NNI header

NNI does not include the Generic Flow Control field NNI header has a Virtual Path Identifier (VPI) field that occupies the first 12 bits, allowing larger trunks between public ATM switches

Fields in the ATM cell header

Generic Flow Control (GFC)

Provides local functions, such as identifying multiple stations that share a single ATM interface

Virtual Path Identifier (VPI)

In conjunctions with the VCI, identifies the next destination of a cell as it passes through a series of ATM switches on the way to the destination

Virtual Channel Identifier (VCI)

In conjunction with the VPI, identifies the next destination of a cell as it passes through a series of ATM switches on the way to the destination

Payload Type (PT)

Indicates in the first bit whether the cell contains user data or control data; if the cell contains user data, the second bit indicates congestion, and the third bit indicates whether the cell is the last in a series of cells that represent a single AAL5 frame

Cell Loss Priority (CLP)

Indicates whether the cell should be discarded if there is congestion in the network; if the CLP bit equals 1, the cell should be discarded in preference to cells with the CLP bit equal to zero

Header Error Control (HEC)

Calculates the checksum only on the header itself; if any cell that fails the header error check is instantly discarded by the network

ATM connection

ATM is a connection-oriented technique

The end to end route is defined through the network in the beginning of the connection and that remains the same through the connection

ATM cells are routed on the same route to both directions; this guarantees that the cells arrive in the receiving end in the same order that they are sent

Also cell delay variation is minimized

ATM virtual connections

Virtual connections are used to provide connectivity between communicating endpoints There are two types of ATM connections:

Virtual Channel Connection (VCC) Virtual Path Connection (VPC)

Each ATM cell contains a label in its header to explicitly identify the VC to which the cell belongs; this label consist of two parts:

Virtual Channel Identifier (VCI) Virtual Path Identifier (VPI)

Virtual Channel Connection (VCC) is a logical connection in ATM Virtual Channel Identifier (VCI) identifies a particular VC link under a given VPC; a specific value of a VCI is assigned each time a VC is switched in the network Virtual Path Connection (VPC) is a local grouping of VCC having the same endpoint; al the cells flowing in a single VPC are switched together; also virtual paths are used for bundling a number of virtual channels into a higher bandwidth stream routed through ATM switches Virtual Path Identifier (VPI) identifies a group of VC links at a given reference point that share the same VPC; a specific value of a VPI is assigned each time a VP is switched in the network

Transmission Path is a bundle of VPs.

Virtual Path Connection (VPC) advantages

Simplified network architecture

Network transport functions can be separated into those related to individual logical connections (VCC) and those related to a group pf logical connections (VPC)

Increased network performance and reliability

The network deals with fewer, aggregated entities

Segregation of traffic

A form of priority control can be implemented by separating traffic types requiring different quality of service

Reduced processing and short connection setup time

By reserving capacity on a VPC in anticipation of later call arrivals, new VCCs can be established by executing simple control functions at the end points of the VPC; no call processing is required at transit nodes

Enhanced network Services

The VPC is used internally in the network but is also visible to the end user; the user may define closed user groups or closed networks of VC bundles

ATM traffic management

ATM provides a mechanism to ensure that the quality of service remains as high as possible while the operator is able to utilize the network capacity in an efficient way

The actions taken to achieve the QoS targets in ATM networks are called traffic management; the two basic control functions are:

Connection Admission Control (CAC)

Used for checking that there are bandwidth and buffer resources for requested connections; CAC is defined as a set of actions taken by the network at the connection establishment phase in order to decide whether a virtual channel/path connection can be accepted or rejected; CAC verifies whether network (or node) is able to offer QoS without risking the QoS of the existing connections

Usage/Network Parameter Control (UPC/NPC)

Also know as traffic policing, is used for monitoring the compliance of ATM end systems to agreed traffic contracts Traffic contacts defines traffic parameters (e.g. peak cell rate) for each connection UPC is applied for User-Network Interface (UNI) NPC is used for Network-Node Interface (NNI) Traffic policing is define as a set of actions taken by the network to monitor and control the amount of incoming ATM traffic The main purpose is to protect resources from misbehavior that can affect the QoS of other already established connections It is done by detecting violations of negotiated parameters and taking appropriate action:

Cell passing Cell tagging Cell discarding

Cell priority and frame discard

Cell los priority bit in a ATM cell header can be used to generate different priority cell flows within a virtual path or channel connection A network element may selectively discard cells with low priority before higher priority cell in congestion situation If a congested network needs to discard cells, its better to drop all the cells of one Protocol Data Unit (PDU) rather than to randomly drop cells belonging to different PDUs; if a single cell is discarded, it may cause the retransmission of the whole PDU, which in turn may cause more traffic when congestion is already occurring The most two common congestion control mechanism implemented in ATM are:

Partial Packet Discard

If a cell is dropped from a switch buffer, the subsequent cells in the higher layer protocol datagram are also discarded

Early Packet Discard

When the switch buffer queues reach a threshold level, the entire higher level datagrams are droped

Traffic contract and negotiation

Traffic contract negotiated during connection establishment consists of the connection traffic descriptor, the requested QoS class and the definition of a compliant connection

The values of the traffic contract parameters can be specified either explicitly or implicitly

A parameter value is explicitly specified in the initial call establishment message

A parameter value is implicitly specified when its value is assigned by the network using default rules

ATM quality of service (QoS)

The ATM supports QoS guarantees for different types of traffic There are six QoS parameters, three of them are negotiated between the end system and the networks

The QoS parameters that are negotiated:

Cell Loss Ratio (CLR) Cell Transfer Delay (CTD) Cell Delay Variation (CDV)

The QoS parameters that are not negotiated:

Cell Error Ratio (CER) Severely Error Cell Block Ratio (SECBR) Cell Misinsertion Rate (CMR)

Propagation delay dominates the fixed delay component of CTD, while queuing behavior contributes to delay variations in heavily loaded networks The effects of queue service strategy and buffer sizes dominates loss and delay variation performance in congested networks Transmission link error characteristics largely determine the CER, SECBR and CMR parameters

Traffic descriptors

The traffic parameters describe traffic characteristics of a source The traffic descriptors are the generic list of traffic parameters, which describe the traffic characteristics of an ATM connection:

Peak Cell Rate (PCR)

Represents the maximum allowable cell rate It is defined as the inverse of the minimum inter-arrival time between two consecutive cells (called the peak emission interval of the connection)

Sustained Cell Rate (SCR)

Represents a theoretical average of the cell average rate (cell) transmission sustained over the duration of a transmission for a VBR connection

Burst Tolerance (BT)

Represents the maximum time in advance allowed to transmit a cell compared to its nominal transmission time

Maximum Burst Size (MBS)

Specifies the maximum number of cells that can be transmitted at PCR while still being compliant to the negotiated SCR

Minimum Cell Rate (MCR)

Represents the minimum users required bandwidth or the minimum cell rate guaranteed for the user (it is the descriptor used for ABR service)

Cell Delay Variation Tolerance (CDVT)

Gives information on the maximum time in advance allowed for a cell arrival (it can be used with PCR and SCR)

ATM Service Categories

Constant Bit Rate (CBR)

Is used by applications that request a static amount of bandwidth that is continuously available during the connection lifetime This amount of bandwidth is characterized by a peak cell rate value The source can emit cells at the peak cell rate at any time for any duration and the QoS commitments still pertain

Real time Variable Bit Rate (rt-VBR)

Service category intended for real time applications that transmit at a rate that varies over time Connections are categorized in terms of peak cell rate, sustainable cell rate and maximum burst size Cells that are delayed beyond the value specified by maxCTD are assumed to be of significantly reduced value to the application

Non-real Time Variable Bit Rate (nrt-VBR)

Service category intended for non-real time application which have bursty traffic characteristics It has same traffic parameters as the rt-VBR No delay bounds are associated with this service category

Available Bit Rate (ABR)

Indented for non real time applications sources having ability to reduce or increase their information rate if the network requires them to do so A flow control mechanism is specified to provide feedback to control the source rate No specific QoS parameter is negotiated

Unspecified Bit Rate (UBR)

Intended for non-real time applications that do not have strict requirements for delay and delay variance Does not have guaranteed QoS and is also known as best effort traffic class

Access profile of ATM interface

The access profile created for an ATM interface defines the connection structure built under that ATM interface and the following parameters must be specified

Interface id The maximum ingress and egress transmission bandwidth The maximum ingress and egress bandwidth unit The maximum number of VPI bits The maximum numbers of VCI bits

ATM Protocol

The ATM reference model includes three planes, which consist of layers:

User Plane is responsible for user information transfer and associated controls (flow control and error control) Control lane performs call and connection control functions

Management plane contains two components:

Layer management (which performs management functions relating to layers resources and parameters) Plane management (which performs managements functions related to the system as a whole)

The ATM protocol reference model includes three functional layers:

Physical layer (which defines the transmission medium, electrical characteristic, network interfaces and a signal-encoding scheme and is further divided)

Physical medium dependent sub layer (responsible for coding, encoding, scrambling and adaptation to the medium; ATM can use any physical medium capable of carrying ATM cells such as SDH, SONET and E1) Transmission convergence sub layer (handles all the processes involved in taking cells to/from the ATM layer and performs bit rate adaptation, header protection, cell degradation and adaptation to the physical mediums structure)

ATM layer

Adds the cell header to the 48-byte cell payload after it has been assembled in the ATM adaptation layer (AAL) and extracts the header before the cell is delivered to the AAL The layer translates the value of the VCI and VPI at the ATM switches or cross-connects It provides virtual connections between endpoint and maintains the contracted QoS by applying a traffic contracted procedure at a call setup time

ATM adaptations layer (AAL)

Provides data link services for the upper layer protocols Needed for adapting upper-layer protocol data units such as TCP/IP and signaling to ATM layer Furthermore voice codecs generate short voice packets which must be adapted to ATM layer services

ATM Adaptation Layer

AAL layer includes two sub layer:

Convergence sub layer (CS)

Provides the AAL service to the higher layer protocol This sub layer is service independent It performs a variety of functions that depend on the actual service being supported including clock recovery, delay variation and dealing with other problems introduced by the network

Segmentation and assembly sub layer (SAR)

Provides segmentation of the users information into 48-byte segments that form the payload field of an ATM cell Reassembles the content of the ATM cell information fields into higher layer information formats

ATM adaptation layer protocols are handled only at he endpoints of the ATM connection; typically AAL type is transparent for intermediate ATM switches

ATM Adaptation Layer

AAL1

is for constant bit rate (CBR) information, which is requires timing synchronization between the source and destination Appropriate for transporting telephone traffic, uncompressed video traffic and circuit emulation service

ALL2

Used for variable bit rate (VBR) information, which requires a strict relationship between the transmission and reception blocks It provides the bandwidth efficient transmission of short, variable length packets in delay sensitive applications Multiplexes short packets from multiple users into one ATM connection Designed for transporting compressed voice in mobile networks but also compressed voice in wireline applications The AAL is aimed at compressed video, which will vary its bit rate significantly

ALL3/4

Is used for data transmission in a connection oriented or connectionless mode Aimed at variable bit rate information which has no strict timing relationship between the transmitter and receiver

ALL5

Supports connection oriented or connectionless variable bit rate data No timing relationship is required between the transmitter and receiver Also known as Simple and Efficient Adaptation layer (SEAL) Provides similar data transport service to ALL3/4 but it provides the service in a much simpler way and with significantly fewer overheads and it does not include a multiplexing capacity

Switching in ATM network

In ATM networks, the user traffic can be switched in ATM layer or ATM adaptation layer; in mobile networks ALL type 2 switching is used Switching in ATM layer There are two levels of switching capability within an ATM switch (at ATM layer) Virtual Path (VP) switching

Used only when VPI field within the cell header is used to describe the destination of the cells Has the advantage that many VCIs destined for the same network endpoint can be bulk switched VP switches terminate VP links A VP switch translates incoming VPIs to the corresponding outgoing VPIs according to the destination of the VPC whereas VCI values remain unchanged

Virtual Channel (VC) switching

Takes place when all cells on a physical interface are identified and switched to their destination through the switch fabric based on a combination of the VPI/VCI values. A table is maintained on each interface identifying input and output ports associated with certain VPI/VCI The VCI values are changed in a VC switch and the VPI values are changes as they pass through a VP switch; however VCI values are not changed when passing through a VP switch

ALL type 2 switching

AAL2 supports multiplexing of different sources on a single ATM virtual connection The channel identifier (CID) is used to distinguish AAL connections within a single ATM VCC An AAL2 switching system performs AAl2 level switching while an ATM node performs only ATM level switching The traditional VPI/VCI table used for ATM cell switching is extended one more level by introducing CID entries to identify AAL type 2 connections ATM cells received at an AAL type 2 switch are first demultiplexed into AAL type 2 connections (CIDs) then switched and assembled into outgoing ATM cells according to entries found in the VPI/VCI/CID table If an AAl2 connection is routed through ATM switches that do not support AAL2 switching it is considered to be an AAL2 trunking and those switches support only ATM level switching