Control Plane Overview Subnet 1.2 Subnet 1.2

Control Plane Overview Subnet 1.2 Subnet 1.2

The Network Layer: Control Plane Overview Subnet 1.2 Subnet 1.2 R3 R2 R6 Interior Subnet 1.2 R5 Subnet 1.2 Subnet 1.2 R7 1. Routing Algorithms: Link-State, Distance Vector R8 R4 R1 Subnet 1.2 Dijkstra’s algorithm, Bellman-Ford Algorithm Exterior Subnet 1.2 Subnet 1.2 2. Routing Protocols: OSPF, BGP Raj Jain 3. SDN Control Plane Washington University in Saint Louis 4. ICMP Saint Louis, MO 63130 5. SNMP [email protected] Audio/Video recordings of this lecture are available on-line at: Note: This class lecture is based on Chapter 5 of the textbook (Kurose and Ross) and the figures provided by the authors. http://www.cse.wustl.edu/~jain/cse473-16/ Washington University in St. Louis http://www.cse.wustl.edu/~jain/cse473-16/ ©2016 Raj Jain Washington University in St. Louis http://www.cse.wustl.edu/~jain/cse473-16/ ©2016 Raj Jain 5-1 5-2 Network Layer Functions Overview Routing Algorithms T Forwarding: Deciding what to do with a packet using 1. Graph abstraction a routing table Data plane 2. Distance Vector vs. Link State T Routing: Making the routing table Control Plane 3. Dijkstra’s Algorithm 4. Bellman-Ford Algorithm Washington University in St. Louis http://www.cse.wustl.edu/~jain/cse473-16/ ©2016 Raj Jain Washington University in St. Louis http://www.cse.wustl.edu/~jain/cse473-16/ ©2016 Raj Jain 5-3 5-4 Rooting or Routing Routeing or Routing T Rooting is what fans do at football games, what pigs T Routeing: British do for truffles under oak trees in the Vaucluse, and T Routing: American what nursery workers intent on propagation do to T Since Oxford English Dictionary is much heavier than cuttings from plants. any other dictionary of American English, British T Routing is how one creates a beveled edge on a table English generally prevails in the documents produced top or sends a corps of infantrymen into full scale, by ISO and CCITT; wherefore, most of the disorganized retreat international standards for routing standards use the routeing spelling. Ref: Piscitello and Chapin, “Open Systems Networking: TCP/IP and OSI,” Adison-Wesley, 1993, p413 Ref: Piscitello and Chapin, “Open Systems Networking: TCP/IP and OSI,” Adison-Wesley, 1993, p413 Washington University in St. Louis http://www.cse.wustl.edu/~jain/cse473-16/ ©2016 Raj Jain Washington University in St. Louis http://www.cse.wustl.edu/~jain/cse473-16/ ©2016 Raj Jain 5-5 5-6 Graph abstraction Distance Vector vs. Link State 5 Distance Vector: T Vector of distances to all nodes, e.g., 5 T Graph: G = (N,E) v 3 w 2 5 u: {u:0, v:2, w:5, x:1, y:2, z:4} T N = Set of routers v 3 w u 2 1 z T Sent to neighbors, e.g., 5 = { u, v, w, x, y, z } 3 2 1 2 u will send to v, w, x u 2 1 z T x y 3 E = Set of links 1 T Large vectors to small # of nodes 1 x y 2 ={ (u,v), (u,x), (v,x), (v,w), (x,w), (x,y), (w,y), (w,z), (y,z) } Tell about the world to neighbors 1 T Each link has a cost, e.g., c(w,z) = 5 T Older method. Used in RIP. T Cost of path (x1, x2,…, xp) = c(x1,x2) + c(x2,x3) + … + c(xp-1,xp) Link State: T Routing Algorithms find the least cost path T Vector of link cost to neighbors, e.g, u: {v:2, w:5, x:1} T We limit to “Undirected” graphs, i.e., cost is same in both T Sent to all nodes, e.g., u will send to v, w, x, y, z directions T Small vectors to large # of nodes Tell about the neighbors to the world T Newer method. Used in OSPF. Washington University in St. Louis http://www.cse.wustl.edu/~jain/cse473-16/ ©2016 Raj Jain Washington University in St. Louis http://www.cse.wustl.edu/~jain/cse473-16/ ©2016 Raj Jain 5-7 5-8 Dijkstra’s Algorithm Dijkstra’s Algorithm: Example T Goal: Find the least cost paths from a given node to 5 all other nodes in the network v 3 w T Notation: 2 5 c(i,j) = Link cost from i to j if i and j are connected u 2 1 3 z D(k) = Total path cost from s to k 1 x y 2 N’ = Set of nodes so far for which the least cost path is known 1 T Method: T Initialize: N’={u}, D(v) = c(u,v) for all neighbors of u N’ D(v) Path D(w) Path D(x) Path D(y) Path D(z) Path T Repeat until N includes all nodes: 0 {u} 2 u-v 5 u-w 1 u-x f - f - 1 {u, x} 2 u-v 4 u-x-w 2 u-x-y f - Find node w N’, whose D(w) is minimum 2 {u, x, y} 2 u-v 3 u-x-y-w 4 u-x-y-z Add w to N’ 3 {u, x, y, v} 3 u-x-y-w 4 u-x-y-z 4 {u, x, y, v, w} 4 u-x-y-z Update D(v) for each neighbor of w that is not in N’ D(v) = min[D(v), D(w) + c(w,v)] for all v N’ 5 {u, x, y, v, w, z} Washington University in St. Louis http://www.cse.wustl.edu/~jain/cse473-16/ ©2016 Raj Jain Washington University in St. Louis http://www.cse.wustl.edu/~jain/cse473-16/ ©2016 Raj Jain 5-9 5-10 Complexity and Oscillations Homework 5A T Algorithm complexity: n nodes Prepare the routing calculation table for node 1 in the T Each iteration: need to check all nodes, w, not in N following network using Dijkstra’s algorithm T n(n+1)/2 comparisons: O(n2) T More efficient implementations possible: O(n log n) T Oscillations Possible: e.g., support link cost equals amount of 3 carried traffic 2 3 2 1 1 1 1 6 4 4 1 4 5 Washington University in St. Louis http://www.cse.wustl.edu/~jain/cse473-16/ ©2016 Raj Jain Washington University in St. Louis http://www.cse.wustl.edu/~jain/cse473-16/ ©2016 Raj Jain 5-11 5-12 Bellman-Ford Algorithm node x table Bellman Ford Example 1 cost too costcost toto costcost toto T Notation: x y z x y z x y z u = Source node x 0 2 7 x 0 2 3 x 0 2 3 y y 2 0 1 y from from 2 0 1 c(i,j) = link cost from i to j z z from 7 1 0 z 3 1 0 node y table h = Number of hops being considered cost to cost to cost to x y z x y z y Du(n) = Cost of h-hop path from u to n x y z 2 1 x x 0 2 7 T Method: x 0 2 3 x z y 2 0 1 y 2 0 1 y 7 from 2 0 1 from 1. Initialize: D (n) = f for all n z u; D (u) = 0 from z z 7 1 0 z 3 1 0 u u node z table cost to cost to cost to 2. For each node: Du(n) = minj [Du(j) +c(j,n)] x y z x y z x y z 3. If any costs change, repeat step 2 x x 0 2 7 x 0 2 3 y y 2 0 1 y 2 0 1 from from from z z z 71 0 3 1 0 3 1 0 Washington University in St. Louis http://www.cse.wustl.edu/~jain/cse473-16/ ©2016 Raj Jain Washington University in St. Louis http://www.cse.wustl.edu/~jain/cse473-16/ time ©2016 Raj Jain 5-13 5-14 Bellman-Ford Example 2 Bellman-Ford: Tabular Method A.w B. w w 5 v 5 v 5 v f 3 f 5 2 3 5 5 3 5 0 2 f 0 2 f 2 2 1 2 1 2 1 u 3 z u 3 z u 3 z 1 2 1 2 1 2 f 1 f 1 1 f 1 xy xy If cost changes xy C.w D. w 5 v 5 v Recompute the costs to all neighbors 2 4 2 3 3 5 3 5 h D(v) Path D(w) Path D(x) Path D(y) Path D(z) Path 0 2 10 0 2 4 0 f - f - f - f - f - 2 1 2 1 1 2 u-v 5 u-w 1 u-x f - f - uu3 z 3 z 1 2 1 2 2 2 u-v 4 u-x-w 1 u-x 2 u-x-y 10 u-w-z 1 1 2 1 1 2 3 2 u-v 3 u-x-y-w 1 u-x 2 u-x-y 4 u-x-y-z 4 2 u-v 3 u-x-y-w 1 u-x 2 u-x-y 4 u-x-y-z xy xy Washington University in St. Louis http://www.cse.wustl.edu/~jain/cse473-16/ ©2016 Raj Jain Washington University in St. Louis http://www.cse.wustl.edu/~jain/cse473-16/ ©2016 Raj Jain 5-15 5-16 Counting to Infinity Problem Routing Algorithms: Summary A R1 R2 Time 0A 1R1 1.

View Full Text

Details

  • File Type
    pdf
  • Upload Time
    -
  • Content Languages
    English
  • Upload User
    Anonymous/Not logged-in
  • File Pages
    14 Page
  • File Size
    -

Download

Channel Download Status
Express Download Enable

Copyright

We respect the copyrights and intellectual property rights of all users. All uploaded documents are either original works of the uploader or authorized works of the rightful owners.

  • Not to be reproduced or distributed without explicit permission.
  • Not used for commercial purposes outside of approved use cases.
  • Not used to infringe on the rights of the original creators.
  • If you believe any content infringes your copyright, please contact us immediately.

Support

For help with questions, suggestions, or problems, please contact us