Network layer functions

Transcription

1 Network layer functions transport packet from sending to receiving hosts network layer protocols in every host, router application transport network data link physical network data link physical network data link physical network data link physical three important functions: path determination: route taken by packets from source to dest. Routing algorithms switching: move packets from router s input to appropriate router output call setup: some network architectures require router call setup along path before data flows network data link physical network data link physical network data link physical network data link physical network data link physical application transport network data link physical 4: Network Layer 4a-1

2 Network service model service abstraction Q: What service model for channel transporting packets from sender to receiver? guaranteed bandwidth? preservation of inter-packet timing (no jitter)? loss-free delivery? in-order delivery? congestion feedback to sender? The most important abstraction provided by network layer: virtual circuit or datagram? 4: Network Layer 4a-2

3 Virtual circuits source-to-dest path behaves much like telephone circuit performance-wise network actions along source-to-dest path call setup, teardown for each call before data can flow each packet carries VC identifier (not destination host OD) every router on source-dest path s maintain state for each passing connection transport-layer connection only involved two end systems link, router resources (bandwidth, buffers) may be allocated to VC to get circuit-like perf. 4: Network Layer 4a-3

4 Virtual circuits: signaling protocols used to setup, maintain teardown VC used in ATM, frame-relay, X.25 not used in today s Internet application transport network data link physical 5. Data flow begins 6. Receive data 4. Call connected 3. Accept call 1. Initiate call 2. incoming call application transport network data link physical 4: Network Layer 4a-4

5 Datagram networks: the Internet model no call setup at network layer routers: no state about end-to-end connections no network-level concept of connection packets typically routed using destination host ID packets between same source-dest pair may take different paths application transport network data link physical 1. Send data 2. Receive data application transport network data link physical 4: Network Layer 4a-5

6 Network layer service models: Network Architecture Service Model Bandwidth Guarantees? Loss Order Timing Congestion feedback Internet ATM ATM ATM ATM best effort CBR VBR ABR UBR none constant rate guaranteed rate guaranteed minimum none no yes yes no no no yes yes yes yes no yes yes no no no (inferred via loss) no congestion no congestion yes no Internet model being extented: Intserv, Diffserv Chapter 6 4: Network Layer 4a-6

7 Datagram or VC network: why? Internet data exchange among computers elastic service, no strict timing req. smart end systems (computers) can adapt, perform control, error recovery simple inside network, complexity at edge many link types different characteristics uniform service difficult ATM evolved from telephony human conversation: strict timing, reliability requirements need for guaranteed service dumb end systems telephones complexity inside network 4: Network Layer 4a-7

8 Routing Routing protocol Goal: determine good path (sequence of routers) thru network from source to dest. Graph abstraction for routing algorithms: graph nodes are routers graph edges are physical links link cost: delay, $ cost, or congestion level A B D good path: 1 typically means minimum cost path other def s possible C E F 4: Network Layer 4a-8

9 Routing Algorithm classification Global or decentralized information? Global: all routers have complete topology, link cost info link state algorithms Decentralized: router knows physicallyconnected neighbors, link costs to neighbors iterative process of computation, exchange of info with neighbors distance vector algorithms Static or dynamic? Static: routes change slowly over time Dynamic: routes change more quickly periodic update in response to link cost changes 4: Network Layer 4a-9

10 A Link-State Routing Algorithm Dijkstra s algorithm net topology, link costs known to all nodes accomplished via link state broadcast all nodes have same info computes least cost paths from one node ( source ) to all other nodes gives routing table for that node iterative: after k iterations, know least cost path to k dest. s Notation: c(i,j): link cost from node i to j. cost infinite if not direct neighbors D(v): current value of cost of path from source to dest. V p(v): predecessor node along path from source to v, that is next v N: set of nodes whose least cost path definitively known 4: Network Layer 4a-10

11 Dijsktra s Algorithm 1 Initialization: 2 N = {A} 3 for all nodes v 4 if v adjacent to A 5 then D(v) = c(a,v) 6 else D(v) = infty 7 8 Loop 9 find w not in N such that D(w) is a minimum 10 add w to N 11 update D(v) for all v adjacent to w and not in N: 12 D(v) = min( D(v), D(w) + c(w,v) ) 13 /* new cost to v is either old cost to v or known 14 shortest path cost to w plus cost from w to v */ 15 until all nodes in N 4: Network Layer 4a-11

12 Dijkstra s algorithm: example Step start N A AD ADE ADEB ADEBC ADEBCF D(B),p(B) 2,A 2,A 2,A D(C),p(C) 5,A 4,D 3,E 3,E D(D),p(D) 1,A D(E),p(E) infinity 2,D D(F),p(F) infinity infinity 4,E 4,E 4,E 5 A 1 2 B D C E F 4: Network Layer 4a-12

13 Dijkstra s algorithm, discussion Algorithm complexity: n nodes each iteration: need to check all nodes, w, not in N n*(n+1)/2 comparisons: O(n**2) more efficient implementations possible: O(nlogn) Oscillations possible: e.g., link cost = amount of carried traffic 1 D A 1 1+e e C e initially B 1 D A 2+e e 1 C 0 B recompute routing D A 0 2+e e C B recompute D A 2+e e 1 C e recompute B 4: Network Layer 4a-13

14 Distance Vector Routing Algorithm iterative: continues until no nodes exchange info. self-terminating: no signal to stop asynchronous: nodes need not exchange info/iterate in lock step! distributed: each node communicates only with directly-attached neighbors Distance Table data structure each node has its own row for each possible destination column for each directlyattached neighbor to node example: in node X, for dest. Y via neighbor Z: X D (Y,Z) = = distance from X to Y, via Z as next hop Z c(x,z) + min {D (Y,w)} w 4: Network Layer 4a-14

15 Distance Table: example A 1 7 E D (C,D) E D (A,D) E D (A,B) B E C D D = c(e,d) + min {D (C,w)} w = 2+2 = 4 D = c(e,d) + min {D (A,w)} w = 2+3 = 5 2 loop! = c(e,b) + min {D (A,w)} w = 8+6 = 14 loop! B destination E D () A B C D cost to destination via A B D : Network Layer 4a-15

16 Distance table gives routing table E D () cost to destination via A B D Outgoing link to use, cost A A A,1 destination B C destination B C D,5 D,4 D D D,4 Distance table Routing table 4: Network Layer 4a-16

17 Distance Vector Routing: overview Iterative, asynchronous: each local iteration caused by: local link cost change message from neighbor: its least cost path change from neighbor Distributed: each node notifies neighbors only when its least cost path to any destination changes neighbors then notify their neighbors if necessary Each node: wait for (change in local link cost of msg from neighbor) recompute distance table if least cost path to any dest has changed, notify neighbors 4: Network Layer 4a-17

18 Distance Vector Algorithm: At all nodes, X: 1 Initialization: 2 for all adjacent nodes v: 3 D X(*,v) = infty /* the * operator means "for all rows" */ X 4 D (v,v) = c(x,v) 5 for all destinations, y X 6 send min D (y,w) to each neighbor /* w over all X's neighbors */ w 4: Network Layer 4a-18

19 Distance Vector Algorithm (cont.): 8 loop 9 wait (until I see a link cost change to neighbor V 10 or until I receive update from neighbor V) if (c(x,v) changes by d) 13 /* change cost to all dest's via neighbor v by d */ 14 /* note: d could be positive or negative */ 15 for all destinations y: D X(y,V) = D X(y,V) + d else if (update received from V wrt destination Y) 18 /* shortest path from V to some Y has changed */ 19 /* V has sent a new value for its min w DV(Y,w) */ 20 /* call this received new value is "newval" */ 21 for the single destination y: D X (Y,V) = c(x,v) + newval 22 X 23 if we have a new min w D (Y,w)for any destination Y X 24 send new value of min w D (Y,w) to all neighbors forever 4: Network Layer 4a-19

20 Distance Vector Algorithm: example X 2 Y 1 7 Z 4: Network Layer 4a-20

21 Distance Vector Algorithm: example X 2 Y 1 7 Z X D (Y,Z) = c(x,z) + min {D (Y,w)} w = 7+1 = 8 Z Y X D (Z,Y) = c(x,y) + min {D (Z,w)} w = 2+1 = 3 4: Network Layer 4a-21

22 Distance Vector: link cost changes Link cost changes: node detects local link cost change updates distance table (line 15) if cost change in least cost path, notify neighbors (lines 23,24) 1 X 4 Y 50 1 Z good news travels fast algorithm terminates 4: Network Layer 4a-22

23 Distance Vector: link cost changes Link cost changes: good news travels fast bad news travels slow - count to infinity problem! 60 X 4 Y 50 1 Z algorithm continues on! 4: Network Layer 4a-23

24 Distance Vector: poisoned reverse If Z routes through Y to get to X : Z tells Y its (Z s) distance to X is infinite (so Y won t route to X via Z) will this completely solve count to infinity problem? 60 X 4 Y 50 1 Z algorithm terminates 4: Network Layer 4a-24

25 Comparison of LS and DV algorithms Message complexity LS: with n nodes, E links, O(nE) msgs sent each DV: exchange between neighbors only convergence time varies Speed of Convergence LS: O(n**2) algorithm requires O(nE) msgs may have oscillations DV: convergence time varies may be routing loops count-to-infinity problem Robustness: what happens if router malfunctions? LS: DV: node can advertise incorrect link cost each node computes only its own table DV node can advertise incorrect path cost each node s table used by others error propagate thru network 4: Network Layer 4a-25

26 Hierarchical Routing Our routing study thus far - idealization all routers identical network flat not true in practice scale: with 50 million destinations: can t store all dest s in routing tables! routing table exchange would swamp links! administrative autonomy internet = network of networks each network admin may want to control routing in its own network 4: Network Layer 4a-26

27 Hierarchical Routing aggregate routers into regions, autonomous systems (AS) routers in same AS run same routing protocol inter-as routing protocol routers in different AS can run different inter- AS routing protocol gateway routers special routers in AS run inter-as routing protocol with all other routers in AS also responsible for routing to destinations outside AS run intra-as routing protocol with other gateway routers 4: Network Layer 4a-27

28 Intra-AS and Inter-AS routing a C C.b b d A A.a a b A.c c B.a a B c Gateways: perform inter-as routing amongst themselves b perform intra-as routers with other routers in their AS inter-as, intra-as routing in gateway A.c network layer link layer physical layer 4: Network Layer 4a-28

29 Intra-AS and Inter-AS routing a Host h1 C C.b b A.a Inter-AS routing between A and B A.c a d A b c Intra-AS routing within AS A B.a a B c b Host h2 Intra-AS routing within AS B We ll examine specific inter-as and intra-as Internet routing protocols shortly 4: Network Layer 4a-29

30 The Internet Network layer Host, router network layer functions: Transport layer: TCP, UDP Network layer Routing protocols path selection RIP, OSPF, BGP routing table IP protocol addressing conventions datagram format packet handling conventions ICMP protocol error reporting router signaling Link layer physical layer 4: Network Layer 4a-30

31 IP Addressing IP address: 32-bit identifier for host, router interface interface: connection between host, router and physical link router s typically have multiple interfaces host may have multiple interfaces IP addresses associated with interface, not host, router = : Network Layer 4a-31

32 IP Addressing IP address: network part (high order bits) host part (low order bits) What s a network? (from IP address perspective) device interfaces with same network part of IP address can physically reach each other without intervening router LAN network consisting of 3 IP networks (for IP addresses starting with 223, first 24 bits are network address) 4: Network Layer 4a-32

33 IP Addressing How to find the networks? Detach each interface from router, host create islands of isolated networks Interconnected system consisting of six networks : Network Layer 4a-33

34 IP Addresses class A B C D 0network host 10 network host 110 network host 1110 multicast address 32 bits to to to to : Network Layer 4a-34

35 Getting a datagram from source to dest. IP datagram: misc fields source IP addr dest IP addr data datagram remains unchanged, as it travels source to destination addr fields of interest here A B routing table in A Dest. Net. next router Nhops E 4: Network Layer 4a-35

36 Getting a datagram from source to dest. misc fields Starting at A, given IP datagram addressed to B: data look up net. address of B find B is on same net. as A link layer will send datagram directly to B inside link-layer frame B and A are directly connected A B Dest. Net. next router Nhops E 4: Network Layer 4a-36

37 Getting a datagram from source to dest. misc fields data Starting at A, dest. E: look up network address of E E on different network A, E not directly attached routing table: next hop router to E is link layer sends datagram to router inside linklayer frame datagram arrives at continued.. A B Dest. Net. next router Nhops E 4: Network Layer 4a-37

38 Getting a datagram from source to dest. Arriving at , destined for Dest. next misc fields data network router Nhops interface look up network address of E E on same network as router s interface router, E directly attached link layer sends datagram to inside link-layer frame via interface datagram arrives at !!! (hooray!) A B E 4: Network Layer 4a-38

39 Routing in the Internet The Global Internet consists of Autonomous Systems (AS) interconnected with each other: Stub AS: small corporation Multihomed AS: large corporation (no transit) Transit AS: provider Two-level routing: Intra-AS: administrator is responsible for choice Inter-AS: unique standard 4: Network Layer 4a-39

40 IP datagram format IP protocol version number header length (bytes) type of data max number remaining hops (decremented at each router) upper layer protocol to deliver payload to ver head. len 16-bit identifier time to live type of service upper layer 32 bits flgs length fragment offset Internet checksum 32 bit source IP address 32 bit destination IP address Options (if any) data (variable length, typically a TCP or UDP segment) total datagram length (bytes) for fragmentation/ reassembly E.g. timestamp, record route taken, pecify list of routers to visit. 4: Network Layer 4a-40

41 IP Fragmentation & Reassembly network links have MTU (max.transfer size) - largest possible link-level frame. different link types, different MTUs large IP datagram divided ( fragmented ) within net one datagram becomes several datagrams reassembled only at final destination IP header bits used to identify, order related fragments reassembly fragmentation: in: one large datagram out: 3 smaller datagrams 4: Network Layer 4a-41

42 IP Fragmentation and Reassembly length =4000 ID =x fragflag =0 offset =0 One large datagram becomes several smaller datagrams length =1500 ID =x fragflag =1 offset =0 length =1500 ID =x fragflag =1 offset =1480 length =1040 ID =x fragflag =0 offset =2960 4: Network Layer 4a-42

43 ICMP: Internet Control Message Protocol used by hosts, routers, gateways to communication network-level information error reporting: unreachable host, network, port, protocol echo request/reply (used by ping) network-layer above IP: ICMP msgs carried in IP datagrams ICMP message: type, code plus first 8 bytes of IP datagram causing error Type Code description 0 0 echo reply (ping) 3 0 dest. network unreachable 3 1 dest host unreachable 3 2 dest protocol unreachable 3 3 dest port unreachable 3 6 dest network unknown 3 7 dest host unknown 4 0 source quench (congestion control - not used) 8 0 echo request (ping) 9 0 route advertisement 10 0 router discovery 11 0 TTL expired 12 0 bad IP header 4: Network Layer 4a-43

44 Internet AS Hierarchy 4: Network Layer 4a-44

45 Intra-AS Routing Also known as Interior Gateway Protocols (IGP) Most common IGPs: RIP: Routing Information Protocol OSPF: Open Shortest Path First IGRP: Interior Gateway Routing Protocol (Cisco propr.) 4: Network Layer 4a-45

46 RIP ( Routing Information Protocol) Distance vector type scheme Included in BSD-UNIX Distribution in 1982 Distance metric: # of hops (max = 15 hops) Can you guess why? Distance vector: exchanged every 30 sec via a Response Message (also called Advertisement) Each Advertisement contains up to 25 destination nets 4: Network Layer 4a-46

47 RIP (Routing Information Protocol) Destination Network Next Router Num. of hops to dest. 1 A 2 20 B 2 30 B : Network Layer 4a-47

48 RIP: Link Failure and Recovery If no advertisement heard after 180 sec, neighbor/link dead Routes via the neighbor are invalidated; new advertisements sent to neighbors Neighbors in turn send out new advertisements if their tables changed Link failure info quickly propagates to entire net Poison reverse used to prevent ping-pong loops (infinite distance = 16 hops) 4: Network Layer 4a-48

49 RIP Table processing RIP routing tables managed by an application process called route-d (daemon) Advertisements encapsulated in UDP packets (no reliable delivery required; advertisements are periodically repeated) 4: Network Layer 4a-49

50 RIP Table processing 4: Network Layer 4a-50

51 RIP Table example (continued) RIP Table example (at router giroflee.eurocom.fr): Three attached class C networks (LANs) Router only knows routes to attached LANs Default router used to go up Route multicast address: Loopback interface (for debugging) 4: Network Layer 4a-51

52 RIP Table example Destination Gateway Flags Ref Use Interface UH lo U 2 13 fa U le U 2 25 qaa U 3 0 le0 default UG : Network Layer 4a-52

53 OSPF (Open Shortest Path First) open : publicly available Uses the Link State algorithm LS packet dissemination Topology map at each node Route computation using Dijkstra s alg OSPF advertisement carries one entry per neighbor router Advertisements disseminated to entire Autonomous System (via flooding) 4: Network Layer 4a-53

54 OSPF advanced features (not in RIP) Security: all OSPF messages are authenticated (to prevent malicious intrusion); TCP connections used Multiple same-cost paths allowed (only one path in RIP) For each link, multiple cost metrics for different TOS (eg, satellite link cost set low for best effort; high for real time) Integrated uni- and multicast support: Multicast OSPF (MOSPF) uses same topology data base as OSPF Hierarchical OSPF in large domains. 4: Network Layer 4a-54

55 Hierarchical OSPF 4: Network Layer 4a-55

56 Hierarchical OSPF Two-level hierarchy: local area and backbone. Link-state advertisements do not leave respective areas. Nodes in each area have detailed area topology; they only know direction (shortest path) to networks in other areas. Area Border routers summarize distances to networks in the area and advertise them to other Area Border routers. Backbone routers run an OSPF routing alg limited to the backbone. Boundary routers connect to other ASs. 4: Network Layer 4a-56

57 IGRP (Interior Gateway Routing Protocol) CISCO proprietary; successor of RIP (mid 80s) Distance Vector, like RIP several cost metrics (delay, bandwidth, reliability, load etc) uses TCP to exchange routing updates routing tables exchanged only when costs change Loop-free routing achieved by using a Distributed Updating Alg. (DUAL) based on diffused computation In DUAL, after a distance increase, the routing table is frozen until all affected nodes have learned of the change. 4: Network Layer 4a-57

58 Inter-AS routing 4: Network Layer 4a-58

59 Inter-AS routing (cont) BGP (Border Gateway Protocol): the de facto standard Path Vector protocol: and extension of Distance Vector Each Border Gateway broadcast to neighbors (peers) the entire path (ie, sequence of ASs) to destination For example, Gateway X may store the following path to destination Z: Path (X,Z) = X,Y1,Y2,Y3,,Z 4: Network Layer 4a-59

60 Inter-AS routing (cont) Now, suppose Gwy X send its path to peer Gwy W Gwy W may or may not select the path offered by Gwy X, because of cost, policy ($$$$) or loop prevention reasons. If Gwy W selects the path advertised by Gwy X, then: Path (W,Z) = w, Path (X,Z) Note: path selection based not so much on cost (eg,# of AS hops), but mostly on administrative and policy issues (e.g., do not route packets through competitor s AS) 4: Network Layer 4a-60

61 Inter-AS routing (cont) Peers exchange BGP messages using TCP. OPEN msg opens TCP connection to peer and authenticates sender UPDATE msg advertises new path (or withdraws old) KEEPALIVE msg keeps connection alive in absence of UPDATES; it also serves as ACK to an OPEN request NOTIFICATION msg reports errors in previous msg; also used to close a connection 4: Network Layer 4a-61

62 Why different Intra- and Inter-AS routing? Policy: Inter is concerned with policies (which provider we must select/avoid, etc). Intra is contained in a single organization, so, no policy decisions necessary Scale: Inter provides an extra level of routing table size and routing update traffic reduction above the Intra layer Performance: Intra is focused on performance metrics; needs to keep costs low. In Inter it is difficult to propagate performance metrics efficiently (latency, privacy etc). Besides, policy related information is more meaningful. We need BOTH! 4: Network Layer 4a-62

63 Address Management As Internet grows, we run out of addresses Solution (a): subnetting. Eg, Class B Host field (16bits) is subdivided into <subnet;host> fields Solution (b): CIDR (Classless Inter Domain Routing): assign block of contiguous Class C addresses to the same organization; these addresses all share a common prefix 4: Network Layer 4a-63

64 IP Addressing class A B C D 0network host 10 network host 110 network host 1110 multicast address 32 bits to to to to : Network Layer 4a-64

65 CIDR Classless InterDomain Routing Class A is too large, Class C is too small Everyone has a Class B address!!! Solution: sites are given contiguous blocks of class-c addresses (256 addresses each) and a mask. 4: Network Layer 4a-65

66 CIDR repeated aggregation within same provider leads to shorter and shorter prefixes CIDR helps also routing table size and processing: Border Gwys keep only prefixes and find longest prefix match Also Class-C is partitioned by geography e.g., Europe got to : Network Layer 4a-66

67 Subnetting Detailed in RFC bits 8 bits 8 bits Class B 10 network Subnet host Subnetting splits the host address into two parts Size of each part up to administrator. Only 8 bits to subnet in a class C address! 4: Network Layer 4a-67

68 Subnetting Network ID. Subnet. Host Reduces the size of routing tables. One external class B routing table entry instead of 256 class C addresses. Changes to subnets does not require external announcements LAN : Network Layer 4a-68

69 Subnetting inside a domain Internal routers must be aware of the subnet sizes. Subnet Masks identify the size of subnets. 16 bits 8 bits 8 bits Class B 10 network Subnet host = bit value containing 1s over the network and subnet Ids, and 0s over the host IDs. & with any host to determine subnet number. 4: Network Layer 4a-69

70 Subnet Masks 16 bits 8 bits 8 bits Class B 10 network Subnet host = bits 10 bits 6 bits Class B 10 network Subnet host = Using the high order bits of your IP address, you can determine your class (A,B, or C). 4: Network Layer 4a-70

71 Netstat -nr (unix command) horton> netstat -nr Routing Table: Destination Gateway Flags Ref Use Interface U hme U 3 0 hme0 default UG UH lo0 4: Network Layer 4a-71

72 Subnet Arithmetic Host address subnet mask: & (Class B) this host is on subnet 6 or sometimes written as : Network Layer 4a-72

73 Host routing with Subnets Hosts search routing tables for 1. Matching host address (same LAN) 2. Matching subnet address 3. Matching network address 4. Default route 4: Network Layer 4a-73

74 Address Resolution Protocol (ARP) Interface between Link layer and Network Layer. Allows hosts to query who owns an IP address on the same LAN. Owner responds with hardware address. Allows changes to link layer to be independent of IP addressing. (example to come) 4: Network Layer 4a-74

75 ICMP Redirects ICMP is used for routing error messages TTL expired (traceroute) Host unreachable Echo request (ping program) Also used by default routers to redirect along quicker path. 4: Network Layer 4a-75

76 On-the-same-LAN routing 1. Route lookup determines its on the same subnet. 2. Use ARP to determine what link layer address to send it to. 3. Give it to Link layer LAN : Network Layer 4a-76

77 Through-the-gateway Routing 1. Route lookup determines its on a different subnet -> Go through default route. 2. Use ARP to determine link layer address of gateway Give it to Link layer LAN : Network Layer 4a-77

78 ICMP Redirect Routing 1. Route lookup determines dest. is on different subnet. 2. Use ARP to determine link layer address of gateway. 3. Gateway uses routing tables to determine next hop of Gateway sends ICMP redirect to source LAN Future packets from source routed directly to : Network Layer 4a-78

79 IP version 6 We currently use version 4. v6 changes the packet header, but is incrementally deployable More address 2 128!!!!!!!! Allows better routing table aggregation Headers and options are processed faster Security more integrated is about 3x10 38, which is about 7x10 23 IP addresses per sq. meter of the earth!!!!! 4: Network Layer 4a-79

80 Router Architecture Overview Router main functions: routing algorithms and protocols processing, switching datagrams from an incoming link to an outgoing link Router Components 4: Network Layer 4a-80

81 Input Ports Decentralized switching: perform routing table lookup using a copy of the node routing table stored in the port memory Goal is to complete input port processing at line speed, ie processing time =< frame reception time (eg, with 2.5 Gbps line, 256 bytes long frame, router must perform about 1 million routing table lookups in a second) Queuing occurs if datagrams arrive at rate higher than can be forwarded on switching fabric 4: Network Layer 4a-81

82 Speeding Up Routing Table Lookup Table is stored in a tree structure to facilitate binary search Content Addressable Memory (associative memory), eg Cisco 8500 series routers Caching of recently looked-up addresses Compression of routing tables 4: Network Layer 4a-82

83 Switching Fabric 4: Network Layer 4a-83

84 Switching Via Memory First generation routers: packet is copied under system s (single) CPU control; speed limited by Memory bandwidth. For Memory speed of B packet/sec or pps, throughput is B/2 pps Input Port Memory Output Port System Bus Modern routers: input ports with CPUs that implement output port lookup, and store packets in appropriate locations (= switch) in a shared Memory; eg Cisco Catalyst 8500 switches 4: Network Layer 4a-84

85 Switching Via Bus Input port processors transfer a datagram from input port memory to output port memory via a shared bus Main resource contention is over the bus; switching is limited by bus speed Sufficient speed for access and enterprise routers (not regional or backbone routers) is provided by a Gbps bus; eg Cisco 1900 which has a 1 Gbps bus 4: Network Layer 4a-85

86 Switching Via An Interconnection Network Used to overcome bus bandwidth limitations Banyan networks and other interconnection networks were initially developed to connect processors in a multiprocessor computer system; used in Cisco switches provide up to 60 Gbps through the interconnection network Advanced design incorporates fragmenting a datagram into fixed length cells and switch the cells through the fabric; + better sharing of the switching fabric resulting in higher switching speed 4: Network Layer 4a-86

87 Output Ports Buffering is required to hold datagrams whenever they arrive from the switching fabric at a rate faster than the transmission rate 4: Network Layer 4a-87

88 Queuing At Input and Output Ports Queues build up whenever there is a rate mismatch or blocking. Consider the following scenarios: Fabric speed is faster than all input ports combined; more datagrams are destined to an output port than other output ports; queuing occurs at output port Fabric bandwidth is not as fast as all input ports combined; queuing may occur at input queues; HOL blocking: fabric can deliver datagrams from input ports in parallel, except if datagrams are destined to same output port; in this case datagrams are queued at input queues; there may be queued datagrams that are held behind HOL conflict, even when their output port is available 4: Network Layer 4a-88