PNNI - Private Network to Network Interface Principles Topology - PDF document

PNNI - Private Network to Network Interface • Principles • Topology concepts • Routing Protocols • Topology aggregation • Call setup and routing algorithm 6-1 S38.121/RKa s-01 Private-Network-to-Network Interface (PNNI) is ment for interconnection of private network ATM switches • PNNI includes both a routing and a signalling protocol. • Requirements include scalability, efficiency, QoS support, fault tolerance in case of link and node failures and Interoperability with other protocols. • PNNI-routing, like OSPF routing, is based on network topology information which may be aggregated. PNNI supports hierarchy. • PNNI signalling is inherited from the ATM-Forum UNI signalling. Additions are source routing and crankback. 6-2 S38.121/RKa s-01

Why? Why is PNNI based on source routing? 6-3 S38.121/RKa s-01 In the ATM Forum Model PNNI interconnects private networks Public UNI B-ICI Public Public network network ATM user ATM Private network Private network user or switch or switch PNNI Private UNI B-ICI = B-ISDN Inter Carrier Interface PNNI 1.0 specs is af-pnni-0055.000, updated in march 1996, more than 365 pgs. 6-4 S38.121/RKa s-01

PNNI node reference model Management- Topology protocol Topology protocol Route Topology information calculation database exchange UNI NNI signalling signalling UNI NNI Call signalling signalling control Cell flow Cell flow Switching matrix 6-5 S38.121/RKa s-01 PNNI routing functions include • Finding neighbors, links and link states using the Hello - protocol. Establishment of Peer Groups . • Synchronization of the Topology databases by exchanging PNNI Topology State Elements (PTSEs) horizontally inside a peer group. • Election of Peer Group Leaders (PGL) based on PTSEs. • Aggregation of Topology information (task of PGL). • Building up the routing hierarchy (PGL passes to the parent group an aggregated description of his peer group) 6-6 S38.121/RKa s-01

Let us view the PNNI topology concepts and topology protocols 6-7 S38.121/RKa s-01 Peer Group is the key concept in PNNI routing A.4.1 • Peer Group is a set of logical nodes , such that they have A.4.2 A.4.3 the same topology information. This includes both A.4.4 the information about the group itself as well as the description of the rest of the network. A.4.6 A.4.5 PG(A.4) • Nodes have a common address prefix (e.g. A.4) for the sake of efficient coding. The prefix is a configuration parameter set by the operator. • A reasonable size of a Peer Group is max. tens of nodes (e.g. 30 .... 50). 6-8 S38.121/RKa s-01

An example topology PG(A.2) PG(B.1) PG(A.1) 2 PG(B.2) 1 PG(A.3) 1 1 A.1.3 PG(C) 1 2 2 3 A.1.2 C.2 5 4 2 3 A.1.1 C.1 A.4.1 4 3 A.4.2 A.4.3 A.4.4 A.4.6 A.4.5 PG(A.4) 6-9 S38.121/RKa s-01 Peer Groups form a hierarchy • PGL -peer group leader (cmp. designated router in OSPF) aggregates PG A the description of the group and passes PG(A.2) it up in the hierarchy to the next higher PG(A.1) 2 level peer group. 1 PG(A.3) A.1.3 1 A.1.2 4 2 • PGL also receives external topo- A.1.1 3 logy info and distributes it in its group. • Peer groups form a hierarchy. PG(A.4) Address resolution decreases up i.e. prefix becomes shorter. Length of Prefix tells the level in the hierarchy, numbering of levels starts from the top. 6-10 S38.121/RKa s-01

Topology consists of logical nodes and logical links PG A On upper levels: PG(A.2) • a logical node represents the child PG(A.1) 2 1 peer group. In practice the functions of PG(A.3) A.1.3 1 the logical node are taken care of by A.1.2 4 2 A.1.1 3 the PGL of the child group. • Logical link = direct link connecting child peer groups PG(A.4) In the lowest level Peer Group • logical node = physical node. • logical link = physical link 6-11 S38.121/RKa s-01 Election of Peer Group Leader is largely automatic and does not interfere setting up connections Tasks of the PGL are • to aggregate the group topology description A.4.1 • pass it upwards in the group hierarchy A.4.2 • receive topology information sent by the parent group A.4.3 and distribute it in its group. A.4.4 • PGL can be re-elected automatically without operator interference. A.4.6 A.4.5 • Election of the PGL is based on collected topology PG(A.4) info. • Not all nodes need to be eligible. • To be elected a node needs to have a high enough priority and it must know the identity of the parent group • The priority of the elected PGL is increased for stability 6-12 S38.121/RKa s-01

PNNI Topology State Elements (PTSEs) describe the topology PTSEs are built of data sent by the Hello protocol and distributed into the Peer Groups. PTSE identity and order Header PTSE aging Sender identity Sender Sender routing capability, eligibility and PGL priority information Topology Link (horizontal/vertical) and node parameters: information divided into attributes and metrics Internal and External (also non-PNNI) addresses, Reachability to which the node will route traffic information 6-13 S38.121/RKa s-01 Peer Group topology is aggregated by abstracting its real structure into a logical node Port 3 Exception Spoke with default attributes bypass Spoke with exception attributes Port 1 Port 2 A.4.1 Nucleus A.4.2 A.4.3 Spoke with default attributes A.4.4 A.4.6 A.4.5 Kuva: Abstract representation of logical node A.4. PG(A.4) 6-14 S38.121/RKa s-01

Peer Group Leaders build and maintain group hierarchy • Bottom level PGLs build their parent peer groups (NOTE: children create their parents!!!) • Parent peer group has a consistent topology database • Topology of the parent group is distributed in the child groups • A PGL is elected in the parent group • The PGL of the parent group represents the group in the next upper level parent peer group • Key criteria of group membership is longest common address prefix 6-15 S38.121/RKa s-01 An example hierarchical topology Top PG A PG B PG(A.2) PG(B.1) PG(A.1) 2 PG(B.2) 1 PG(A.3) 1 1 A.1.3 PG(C) 1 2 2 3 A.1.2 C.2 5 4 2 3 A.1.1 C.1 A.4.1 4 3 A.4.2 A.4.3 A.4.4 A.4.6 A.4.5 PG(A.4) 6-16 S38.121/RKa s-01

Hello protocol works on a well defined VCC between neighbors Logical Logical node A node B Hello packet contains ATM End System Address Node ID Port ID of the link Peer Group ID • Hello protocol works continuously and reveals link failures. • Hello protocol data is used to form the initial version of the topology database. 6-17 S38.121/RKa s-01 When neighbors have been identified by Hello protocol, topology databases are synchronized Logical Logical node B node A PTSE-header advertisement New info PTSE-requests Yes PTSEs PTSE-ack(headers) Update the DB PTSE-header advertisement 6-18 S38.121/RKa s-01

PNNI flooding protocol is similar to OSPF-flooding Logical Logical node B All other node A PTSEs neighbors except PTSE-ack(headers) sender Refresh timer New info Remove old info from DB Yes Update DB PTSEs Timer PTSEs Event PTSE-ack(headers) (significant change) • Send frequency of PTSEs is a compromise between probability of misrouting and the need to minimize the amount of PTSE-information. • What is a significant change should be configurable. 6-19 S38.121/RKa s-01 Parent peer groups are similar to lowest level peer groups PG A Logical group node - LGN (logical node) has PG(A.2) • ATM End System Address (a different SEL PG(A.1) 2 1 PG(A.3) than PGL) A.1.3 1 A.1.2 4 2 A.1.1 • VCCs are set up between logical group nodes 3 for communication acc to PNNI • PGL is elected in the parent group as well PG(A.4) • PGL is not needed on the topmost level. 6-20 S38.121/RKa s-01

Border nodes describe connections to neighboring groups as uplinks Upnode A.3 A.4 uplink A.4.6 -- A.3 PG A uplink A.3.4 -- A.4 • Topology data is not synchronized between peer PG(A.3) groups on the same level (e.g. A.4.6 -- A.3.4) 1 4 2 • Border nodes exchange information about the 3 hierarchy using Hello protocol and deduce which is the lowest common peer group A.4.6 PG(A.4) • Uplink is the way of the border node to tell its group about a connection to the parent group • Using uplink info (PGLs)/LGNs can set up VCCs between nodes 6-21 S38.121/RKa s-01 PNNI signaling and routing algorithm 6-22 S38.121/RKa s-01

Designated Transit List is a stack representation of the route e.g. from A.1.1 to C.2 On PG border the used part of the route is removed PG A.1: A.1.1->A.1.2->A.1.1.3 PG A: A.1->A.2->A.3 Top Bottom of DTL A.2 border node extends own PG description Top of the stack (DTL) and adds it to the top of stack DTL pointer is updated in each internal node DTL = designated transit list 6-23 S38.121/RKa s-01 Metrics are additive in route calculations PNNI supports QoS routing/route optimization using metrics: • Cell delay variation (CDV) • Maximum Cell Transfer Delay (maxCTD) • Administrative weight (AW) - administrator can define the interpretation of AW Optimization is done using one metric at a time. 6-24 S38.121/RKa s-01