ATM BASICS - Chapter 5 ppt

AESA also includes support of E.164 address format E.164 ATM to ensure compatibility with classical public networks such as PSTN and ISDN.. The for-mat of a private ATM address is totall

Using ATM to Connect Systems

ATM is commonly used to connect systems There exits also a number of

pro-tocols and scenarios for interworking between ATM networks This chapter

introduces signaling flavors and routing concepts, which are deployed in

contemporary ATM networks In the first place, the chapter covers

address-ing as well as general signaladdress-ing issues Then it presents signaladdress-ing protocols

that are used to setup connections across UNI and PNNI interfaces Finally

different methods of interworking between ATM networks are discussed

Connecting systems using ATM must involve the use of ATM addressing

and signaling Both elements are important parts on the ATM concept

Addressing actually takes place at two levels: at the ATM level using

VPI/VCI identifiers and at the logical network level All signaling protocols

assign match VPI/VCI values to ATM address and physical ATM UNI and

/or NNI ports Every physical ATM UNI port must have at least one unique

ATM address Both VCCs and VPCs are unidirectional, so signaling

proto-cols must specify traffic parameters separately for each direction of the

con-nections Note that in addition to signaling operation there is also a need for

routing capabilities necessary to find the optimal route for a connection

5.1 ATM Address Formats

There are two different types of ATM addressing plans First, ATM supports the use of NANP (North American Numbering Plan) that is based on E.164 ITU-T Recommendation Such addressing format is called E.164-Native or NSAP E.164 format Secondly, ATM supports the alternative addressing plan developed by ATM Forum and called ATM End System Addressing (AESA) The AESA addresses are used in private and enterprise addresses AESA also includes support of E.164 address format (E.164 ATM) to ensure compatibility with classical public networks such as PSTN and ISDN Each ATM end station requires a unique ATM address to uniquely identify other end stations The ATM network also uses unique ATM addresses to locate the destination end switch Finally ATM switches themselves use ATM addresses in PNNI networks to uniquely identify each switch The for-mat of a private ATM address is totally different from addresses used in other technologies The size of an AESA ATM address is equal to 20 bytes, which is 16 bytes more than IPv4 address and 4 bytes more that IPv6 address The structure of an AESA address is given in the Fig 5-1

There are three major parts of an ATM address used within a private ATM

network:

•The 13-bytes Network Prefix field, which is used to identify the

net-work part of the ATM address This field may have different formats

depending on the type of the network and the position in the network

•The 6-bytes Endpoint System Identifier, which is a hardware part

of the ATM address This filed is also referred to as a MAC address, as

it uniquely identifies ATM hardware The 48-bits of ESI are assigned by

the vendor of the ATM equipment

•The 1-byte Selector field that is a logical part of the ATM address

It is used to identify a logical function in the ATM device For example,

an ATM end station may support a number of terminal equipment to

access an ATM network, and the Selector field may be used to address

different devices The field is not used in routing

Note that a particular ATM device may have more than one ATM address

It is possible due to the fact that different values of the Selector field can be

hosen to create separate ATM addresses

Public and private ATM networks use different ATM address formats

Public ATM networks use E.164 addresses (i.e., telephone numbers) Private

ATM network addresses are based on the Open System Interconnection

(OSI) Network Service Access Point (NSAP) format NSAP addresses are

based on the concept of hierarchical addressing domains The exact format

of an ATM address is coded in the first byte of an ATM address This byte is

called the AFI (Address Format Indicator) The three different formats are

given in the Fig.5-2

The fields of the address formats are as follows:

•The DCC (Data Country Code) specifies the country in which the address is registered

•The ICD (International Code Designator) provides unified address-ing for international organizations The ICD code is issued centrally by the British Standards Institute (BSI)

•The E.164 address specifies the numbering system with ISDN In ATM, the international numbering format is used These numbers may con-tain up to 15 figures, the length of the E.164 field is eight bytes

•The DFI (Domain-specific part Format Identifier) specifies the structure of the rest of the address field (the AA, RD, Area, ESI and SEL fields)

Fig 5-2, Different formats of ATM addresses

•The AA (Administrative Authority) field identifies national

organi-zations such as the ATM network operator, ATM user or ATM device

man-ufacturer

•The RD (Routing Domains) field specifies an address range that

must be unique within E.164, DCC/DFI/AA or ICD/DFI/AA

BULThe Area field specifies a unique range of addresses with a routing

domain

As it can be easily derived from any format of the ATM address, their

struc-ture follows a hierarchical model This model is given in the Fig 5-3 The

network operator configures his network device with network prefixes in

way that reflects the topology of his network

Fig 5-3, ATM addressing hierarchy.

It has to be mentioned that some ATM devices in a public ATM network may support also E.164-Native format This format defines an address as an ISDN number or as a telephone number specified by ITU-T

5.1 Address Registration

Address registration is an initial procedure, which is executed when an ATM end station is connected to an ATM switch Therefore, users can move their terminal equipment from one location to another without the need for manual address configuration

The registration of end-station devices is carried out by Integrated Local Management Interface (ILMI), which operates on the object table within a MIB When the ATM interface in the ATM end station is enabled, a cold start trap is transmitted out along reserved VPI 0, VCI 16 The ATM switch receives this start trap and replies with the network prefix associated with that ATM switch by the operator The end station then adds its own MAC part of ATM address and Selector field to the prefix to form a full ATM address This address is sent to the switch where it is registered The end station must be able to accept multiple network prefixes, which enables a prefix to change on-line, if needed, allowing for expansion and interconnec-tion of previously separates networks Also, an ATM switch must be able to accept multiple MAC addresses on the same physical port The network must ignore and transparently handle the one-byte Selector field of the address on an end-to-end basis The SEL field provides a Service Access Point within the end station for multiple functions

5.2 UNI Protocol

The User-Network Interface (UNI) is the point between the end-point ATM equipment and the first ATM switch There have been several versions of the UNI specification, defined by the ATM Forum: UNI 2.0, UNI 3.0, UNI 3.1, UNI 4.0 and UNI 4.1 UNI 2.0 supports only PVCs, while the latter

three versions also support SVCs Note that UNI version 3.1 and later

ver-sions are not backwards compatible with UNI version 3.0 UNI signaling

consists of three layer protocol stack and it largely based on the signaling

protocols developed for narrowband ISDN (ITU-T Q.931) referred to as DSS

1 (Digital Subscriber Signaling No 1) The UNI protocol is aligned with the

ITU-T Q.2931 signaling standard designed for broadband signaling (DSS 2)

UNI signaling uses the predefined VPI 0, VCI 5 for transmission of

signal-ing packets The signalsignal-ing is performed by means of the different signalsignal-ing

messages related to the call establishment; call clearing and operation of

point-to-point connections The UNI messages consist of a variety of basic

building blocks containing the necessary information These building blocks

are known as information elements (IEs) and each element has a standard

4-byte header followed by the IE content Information elements carry

dif-ferent elements such as AAL parameters, the calling party number, traffic

descriptor, the called party number and broadband bearer capability

The process of the SVC establishment is initiated by the Setup message

issued by the ATM user device The signaling message is transported to the

nearest switch via pre-established VC (VPI 0, VCI 5) The Setup message

contains the characteristics (AAL type, traffic descriptor, QoS parameters)

and the destination ATM address Then the switch should perform the CAC

function If the resources in the switch are sufficient to accept a new

con-nection, the switch notifies the edge device that triggered the setup process

with a Call Proceeding message Next the ATM switch reserves resources,

assigns VPI/VCI values and identifies the route to the destination Then the

process of the connection setup takes place internally within the network

involving other signaling protocols Once the device at the destination

sig-nals it is ready, a Connect message is propagated through the network

Each node that receives this message issues a Connect Acknowledgment

Finally, the first ATM switch passes a Connect message to the source

It is important to note that in ATM networks one can find both the instances

of Private UNI as well Public UNI Technically, both interfaces use the same set of equipment Differences are essentially in the service capabilities and

a range of supported features Note that Public UNI can be also used as the interface between private and public ATM network (see the Fig 5-6)

Fig 5-4, The UNI setup process

5.3 PNNI Protocol

Private Network-Network Interface (PNNI) was developed to facilitate the

deployment of large and high performance ATM networks The major intent

was to design a protocol that would overcome limitations related to the use

of the PNNI predecessor – IISP (Interim Inter-switch Signaling Protocol)

IISP was developed to enable small, static routed private ATM network

implementations The greatest effort in PNNI development was to create

the routing algorithms While the routing algorithms were being perfected,

IISP defined how to create static routing tables IISP is not a scalable

solu-tion, as the amount of configuration work grows exponentially with the

number of switches and connections The signaling specification in IISP is

based on Q.2931, with the definition of some new Information Elements

spe-cific to the private network With IISP the signaling process is asymmetric

This means there are a clearly defined user side and a network side

Needless to say, this involves some configuration efforts from the operator

Being based on static routing tables, the switches cannot adapt to changing

topology If a link is down or a switch is down, alternatives will not be

con-sidered unless manually entered in the routing tables Once PNNI 1.0 (in

1997) was issued, the IISP solution to call set-up in the private network was

considered to be far inferior

The decision was to base PNNI signaling UNI 4.0 access signaling

specifi-cation and create a new PNNI routing component However, some further

modifications were necessary in order to avoid asymmetrical approach

imposed by the UNI Finally, the protocol procedures were modified and

PNNI capable devices could operate in peer-to-peer model Without this

modification the contention related to VPI/VCI assignment operation would

be a problematic issue as it was in case of the IISP PNNI has been

primar-ily issued for use in private networks The successful establishment of a

vir-tual circuit requires that an optimal route must be identified with regards

to the customer requirements Hence, the PNNI includes also a dynamic

state routing protocol that implements some of the concepts presents in

clas-sical routing protocols used in IP networks

The PNNI routing protocol is able to collect and distribute link availability information on the facilities between switches within the PNNI network and choose the optimum route The process of the route selection takes into account the type of ATM connection, bandwidth availability, and other QoS parameters The PNNI routing protocol can serve very large private/enter-prises networks due to its hierarchical operation manner On start-up, PNNI nodes send ‘hello’ packets on all interfaces to discover their neighbors

As a part of this process, neighboring nodes exchange their ID numbers Please see the example given in Fig 5-5 All nodes with matching numbers form a logical peer group (PG) using the matching digits as a group identi-fier Nodes with at least one link terminating at a switch in another peer group are considered as Border Nodes Nodes within a group exchange and relay information in the form of PNNI Topology State Packets (PTSPs) about link status including virtual bandwidth, availability and next hop address A reliable transport mechanism is used to ensure that all nodes ultimately share the same database

Fig 5-5, PNNI operation

As part of this process the group members select a Peer Group Leader (PGL)

based on a configured priority number, or by selecting the node with the

low-est ATM address Group leaders low-establish logical connections with each

other and exchange a summary of information about their groups The

net-work, viewed from the perspective of a group leader, will appear to have

much more simplified topology Locally, a PGL will retain the detailed view

of its own group including border nodes and therefore real links to

neigh-boring groups PGLs pass this logical network map to the members of their

own group Each PNNI node, therefore, has a detailed description of its own

group and a logical map on how to get to any other group Note that all

end-station addresses begin with the address sequence that matches the group

to which they are attached

The PNNI application is mainly restricted to the use within a private ATM

network However, PNNI can be also used to connect private networks The

limitations of this model are mainly related to the exchange of topology state

information between private networks own by different administrators

Although the PNNI specification has been issued for use in private

net-works, PNNI proves to be sufficiently scalable and robust to be used in

pub-lic networks Although dynamic routing seems to be an interesting apppub-lica-

applica-tion to the public network operators, in fact the PNNI signaling it is not a

reasonable alternative to CCS network signaling The key point is that

PNNI signaling does not support the transaction-based signaling so

intelli-gent network services such as 800 numbers; mobility and others cannot be

supported Over time, a number of enhancements to PNNI have been

released, thus extending its capabilities They include the support of ABR

traffic category as well as mobility of ATM network and particular

switch-es The most important enhancement is called PNNI Augmented Routing

This standard was developed to allow information about non-ATM services

to be distributed in an ATM network as part of the PNNI topology PAR was

intended to simplify configuration and ongoing operation of IP level routing

protocols when operated over PNNI environment According to PAR

specifi-cation, non-ATM devices such as IP routers equipped with ATM interfaces

can automatically learn about other PAR-capable devices active in the same

ATM network Thanks to this concept the overlay routing network (e.g IP)