link state routing principles
play

Link State Routing Principles The Goal is to avoid the routing loops - PDF document

Link State Routing Principles The Goal is to avoid the routing loops typical of DV routing and to scale to bigger networks and to varying topologies. 5-1 S38.121/RKa s-01 Open Shortest Path First(OSPF) is a recommended link state protocol for


  1. Link State Routing Principles The Goal is to avoid the routing loops typical of DV routing and to scale to bigger networks and to varying topologies. 5-1 S38.121/RKa s-01 Open Shortest Path First(OSPF) is a recommended link state protocol for Interior routing in Internet 1 2 A B C Example network 3 4 5 D E • A Link state protocol maintains the topology map 6 (Link state DB) of the network. • When topology changes, maps are updated quickly. • The map is used to produce the Routing Table. • OSPF is IETF specified link state protocol for Internet - OSPF is recommended as the follower of RIP. 5-2 S38.121/RKa s-01

  2. The map is the full list of all links F rom T o L in k D istance 1 2 A B C A B 1 1 A D 3 1 3 4 5 B A 1 1 B C 2 1 D E B E 4 1 6 C B 2 1 C E 5 1 • One node is responsible for a D A 3 1 D E 6 1 particular entry E B 4 1 • Link directions are separate E C 5 1 entries E D 6 1 5-3 S38.121/RKa s-01 Flooding protocol distributes information about topology changes 1 2 From A to B, link 1, dist = inf, message number A B C xxxxx Flooding Algorithm is 3 4 5 Receive m D E 6 Find entry in link DB L Yes No L.mn = m.mn Yes found L.mn < m.mn No yes No Add entry Update entry Create m from L End Broadcast msg Broadcast msg Send m to all other i/f to all other i/f to sender 5-4 S38.121/RKa s-01

  3. Link DB after distribution of failure of link AB From To Link Distance M-Nr 1 A B 1 inf 2 2 A B C xxxxx A D 3 1 1 B A 1 i in nf f 2 3 4 5 B C 2 1 1 D E B E 4 1 1 6 C B 2 1 1 C E 5 1 1 • Message numbering starts from 1 on D A 3 1 1 node restart. D E 6 1 1 E B 4 1 1 • Modulo arithmetic is used to determine E C 5 1 1 what is “a little bigger than” E D 6 1 1 --> message numbering can overflow without problems. 5-5 S38.121/RKa s-01 If network splits into islands, DBs in islands may diverge 1 2 A B C xxxxx 3 4 5 D xxxxx E 6 From T o L ink D istance M -nr From To Link Distance M -nr A B 1 inf 2 A B 1 inf 2 A D 3 1 1 A D 3 1 1 B A 1 inf 2 B A 1 inf 2 B C 2 1 1 B C 2 1 1 B E 4 1 1 B E 4 1 1 C B 2 1 1 C B 2 1 1 C E 5 1 1 C E 5 1 1 D A 3 1 1 D A 3 1 1 D E 6 1 1 D E 6 inf 2 E B 4 1 1 E B 4 1 1 E C 5 1 1 E C 5 1 1 E D 6 inf 2 E D 6 1 1 5-6 S38.121/RKa s-01

  4. Link 2 fails -> DBs diverge even more 1 2 A B C xxxxx xxxxx DBs at B, C and E: From T o L ink D istance M -nr 3 4 5 A B 1 inf 2 A D 3 1 1 D xxxxx E B A 1 inf 2 6 B C 2 inf 2 B E 4 1 1 C B 2 inf 2 C E 5 1 1 D A 3 1 1 D E 6 1 1 There is no immediate problem, E B 4 1 1 but if link 1 goes up again ... E C 5 1 1 E D 6 inf 2 5-7 S38.121/RKa s-01 After reconnection of the islands “Bringing Up Adjacencies ” is required B A DB description<link-id+m-nr> m.mn>L.mn or m.link-id DB description<link-id+m-nr> not in L Yes Link state request <interesting link-id> Changed information We also talk about DB alignment DB update and flooding or merging of the DBs to all neighbors 5-8 S38.121/RKa s-01

  5. Integrity of the Link DB must be secured • Flooding messages are acknowledged link by link • DB description messages are acknowledged • Each DB entry is protected by obsolescence timer, if an update does not arrive in time, entry is removed. • Each Entry is protected by a checksum • Messages carry also authentication info • But: while update is in progress, some nodes receive info earlier than others --> routing mistakes happen 5-9 S38.121/RKa s-01 OSPF is based on Dijkstra’s SPF algorithm • SPF - shortest path first -algorithm computes the shortest path from source node S to all other nodes • Initially nodes are divided to Evaluated E , the paths from which are known and to other nodes R . • In addition an ordered list of paths O is needed. 5-10 S38.121/RKa s-01

  6. Dijkstra’s shortest-path-first algorithm 1 2 A B C E ={S=B}, R ={A, C, D, E}, P = ∅ 3 4 5 O ={<B,C,1>,<B,E,1>,<B,A,1>} sort D E Return: 6 L O = ∅ or P contains paths F r o m T o L in k D is t. Yes R =unreachable A B 1 1 O(1).m=inf A D 3 1 nodes B A 1 1 B C 2 1 p= <B,C,1> ; O = O \ p; V=p.to-node =C B E 4 1 C B 2 1 V ∈ E C E 5 1 Yes D A 3 1 D E 6 1 E = E ∪ V, R = R \ V E B 4 1 E C 5 1 P = P ∪ p E D 6 1 O = O ∪ < p, L (From=V).To, p.Dist+L.Dist> sort O Converges faster than Bellman-Ford O(M.logM) < O(N.M) NB: this removes loops! 5-11 S38.121/RKa s-01 Advantages of Link State Protocols include • Link State DBs converge quickly, no loops are formed • Metrics can be quite accurate. One protocol can easily support several metrics – Capacity, delay, cost, reliability. • Can maintain several routes to a destination. • Exterior routes can have their own representation. 5-12 S38.121/RKa s-01

  7. Using several metrics requires • Metrics must be stored for each link ( L.et1, L.et2 ... ) • Computing separate Routing Tables for each metric ( P (et1), P (et2) ...) • Link protocol must carry all metrics • User packets must be marked with the required metric. • A Routing loop is possible if different nodes use different metrics for one user packet. 5-13 S38.121/RKa s-01 Spreading load to alternative equidistant paths improves network efficiency 1 2 A B C • Queues in nodes become shorter 3 4 5 • Average delay is decreased • End-to-end jitter decreases D E 6 • Less traffic to reroute under failure conditions – May change packet order because paths may have different delay (queue lengths in nodes) – Difficulty: existing traffic can not be pinned down to primary path so that only overload would take the alternative path --> stability is a problem –When are paths equidistant enough? 5-14 S38.121/RKa s-01

  8. Rule A->Y...->X, if Y.et.X < A.et.X accepts only monotonic alternative routes E ={S=B}, R ={A, C, D, E}, P = ∅ This modified algorithm O ={<B,C,1>,<B,E,1>,<B,A,1>} sort finds also the alternative Return: paths. O = ∅ or P contains paths Yes O(1).m=inf R =unreachable nodes W = (p\V).to-node p= <B,C,1> ; O = O \ p; V=p.to-node =C Yes V ∈ E S.et.W < S.et.V E = E ∪ V, R = R \ V p is a feasible alter- P = P ∪ p native path to V O = O ∪ < p, L (From=V).To, p.Dist+L.Dist> sort O NB: this removes loops! 5-15 S38.121/RKa s-01 Link state protocol can describe several external routes with accurate metrics • DV-protocol capability to describe external routes is limited due to counting to infinity problem and due to complexity of Bellman-Ford algorithm (O(N**2) • Link state protocol is free of those limitations. SPF route computation converges as O(N.logN) - where N = nrof external routes • E.g. if there are 30 000 external routes =>10exp9 vs. 450 000 5-16 S38.121/RKa s-01

  9. OSPF Protocol Principles 5-17 S38.121/RKa s-01 OSPF sees the network as a graph External dest Stub External External ntwrk dest dest OSPF router Transit ntwrk OSPF OSPF router router Stub ntwrk Summary Stub Stub ntwrk ntwrk Stub ntwrk ntwrk 5-18 S38.121/RKa s-01

  10. OSPF makes a difference between a router and a host Uses IP sub-net mask and advertises only a single (sub)net OSPF router Host Host Host A B Z Stub network (aliverkko/sub-net) 5-19 S38.121/RKa s-01 OSPF supports Broadcast networks A B In a Broadcast network • each device can send to each other C D • one can send to all or to a N*(N-1)/2 adjacencies (known neighbors) sub-set of connected devices designated (backup) A B • If it has N routers, they have N*(N-1)/2 adjacencies and • each router would advertise C D N-1 routes to other routers Adjacencies are formed only with the +one stub network Designated router(A) (edusreititin) ==> • A must be selected using the Hello-protocol • Synchronization of Link DBs becomes simpler • Backup designated router (B) should be selected together with the Designated. 5-20 S38.121/RKa s-01

  11. OSPF Flooding Protocol in a Broadcast network Router X, ... Designated Advert (->224.0.0.6 - all designated routers) Distribution (->224.0.0.5 - all OSPF routers) Backup designated stays ==> no need to process acks from all other as silent as possible routers in the sub-net 5-21 S38.121/RKa s-01 In non-BC nets OSPF works in the same way except that Bcasts are replaced by point-to-point messages Designated A B (backup) Router X, ... Designated Advertisement C D Distribution E F Permanent connection with designated Backup designated Permanent connection with backup designated NB: it makes sense to minimize permanent Dial-up connection with connections due to their cost other routers 5-22 S38.121/RKa s-01

Recommend


More recommend