[PPT] - IP MULTICAST Adiseshu Hari, T. V. Lakshman and Gordon Wilfong Nokia PowerPoint Presentation

SLIDE 1

1

Bell Labs

1

IP MULTICAST

Adiseshu Hari, T. V. Lakshman and Gordon Wilfong Nokia Bell Labs DIMACS Workshop on Algorithms for Data Center Networks Rutgers University, NJ

SLIDE 2

2

Bell Labs

Denial-of-Service (DoS) attack amplification
Complex Control Plane
Large Forwarding state
Non aggregable

Nokia 2017

Why is IP Multicast not deployed in public networks?

Multicast state flow Multicast data flow Multicast control plane Multicast data plane

SLIDE 3

3

Bell Labs

Denial-of-Service (DoS) attack amplification
Control state 
Forwarding state

Nokia 2017

Can SDN help with Multicast?

Multicast state flow Multicast data flow Multicast SDN Controller Multicast SDN switch

SLIDE 4

4

Bell Labs

Eliminate unicast forwarding state in SDN:
Path Switching: per-flow routing without per-flow state
New data path suitable for SW switches and programmable packet processors
Encode path in the packet headers
DIMACS 2016

Nokia 2017

Can we eliminate multicast forwarding state in SDN?

SLIDE 5

5

Bell Labs

Nokia 2017

Eliminating unicast forwarding state in SDN using Path Switching

SDN Controller

SLIDE 6

6

Bell Labs

Can we extend Path Switching to encode multicast paths? Can we create an efficient encoding of a multicast path? No blowup in packet size (e.g. using bitmaps) No blowup in storage state (e.g,. encode each multicast tree by a unique identifier)

Nokia 2017

Can we eliminate multicast forwarding state in SDN?

SLIDE 7

7

Bell Labs

Unicast Branching (UB)
Use branching nodes in the network to replicate unicast flows.
Use SDN Flow Table at ingress and egress
Use SDN Group Table at branching nodes
Reduces multicast forwarding state from linear to sublinear in number of forwarding

nodes

Nokia 2017

Can we reduce multicast forwarding state in SDN?

SLIDE 8

8

Bell Labs

Transit Switches First Hop Switches First Hop Switches Endpoints Endpoints Central Controller Branching Nodes With Group Tables Unicast Branching (UB) Reference Diagram

SLIDE 9

9

Bell Labs

Nokia 2017

Reducing multicast forwarding state in SDN using Unicast Branching (UB)

UB state flow UB data flow UB SDN Controller Unicast switch Ingress, Egress, Branching SDN switch

SLIDE 10

10

Bell Labs

Tunable knob to switch between unicast replication and full multicast
Allows for an NFV based implementation
Allows Traffic Engineered branches
Fast Reroute, Per branch QoS
Works at all protocol layers – protocol agnostic
Ethernet, IP, MPLS
Enables unicast only protocols like Segment Routing and TCP to be multicast capable*
HTTP Adaptive Streaming multicast
Efficient content caches
Enables Policy Based Multicast

Nokia 2017

Added advantages of Unicast Branching (UB)

* Requires stateful NFV elements, not just SDN switches for branching points

SLIDE 11

11

Bell Labs

Policy based networking: Rules for non default routing
Geofencing
QoS
Membership filtering
UB enables Policy Based Multicast
Number, location and type of branching nodes

Nokia 2017

Policy Based Multicast

SLIDE 12

12

Bell Labs

Nokia 2017

Where are the Algorithms?

SLIDE 13

13

Bell Labs

Problem 1 definition:
Given an ingress node, a set of egress nodes and a set of branching nodes, build an “optimal”

multicast tree.

What is “optimal”
Usual definition is based on link cost.
Steiner tree problem (NP-complete)

Nokia 2017

Building Efficient Policy Based Multicast Trees

SLIDE 14

14

Bell Labs

UB based multicast tree is not a tree!!!
It is a “configuration”
Cannot apply Steiner tree approximation solutions directly.
Problem: How to create minimum cost configurations?

Nokia 2017

Building Multicast Trees using UB – Major Issue

r

1

r

2

r

G

b

SLIDE 15

15

Bell Labs

Transformation to Steiner tree problem on H

Define :

Edge-weighted graph H = (O,E). O is set of branching nodes (including terminals)
e=(b,b’) ∊ O, w(e) = length shortest path containing no internal O nodes

r

1

r

2

r

5 5 4

H

r

1

r

2

r

G

b

2

Theorem: Minimum cost configuration problem in G is equivalent to Steiner tree problem in H

b

SLIDE 16

16

Bell Labs

Minimum cost configuration problem

Theorem: There is a polynomial-time 1.39-approximation algorithm for min cost configuration problem. [BGRS10] Theorem: The minimum cost configuration problem is APX-hard. Proof: Follows from APX-hardness of Steiner problem for complete graphs with weights 1 and 2. [BP89]

SLIDE 17

17

Bell Labs

Problem 1: Minimize cost given a set of branching nodes . MIN COST PROBLEM
Problem 2: Minimize number of branching nodes given a fixed cost. MIN BRANCHING

PROBLEM

Nokia 2017

Problem 2: Minimize branching nodes

SLIDE 18

18

Bell Labs

Does anybody

Min Branching Problem

For a subset X of the transit nodes, let CX be the minimum cost valid

configuration using X as the set of extra branching nodes.

We are given a graph G = (V, E), a multicast demand d = (r, r1, r2,..., rt),

a bound k and an attainable cost c.

Does there exists a branching set X with least cost

valid configuration CX satisfying d where X  <= k and cost(CX) <= c.

SLIDE 19

19

Bell Labs

Does anybody

Min Branching Problem

Corollary: For this problem the best possible approximation is ≈ ln n . Proof: Follows from bounds for Set Cover. Theorem: This problem is NP-complete. Proof: Follows from a construction using Set Cover.

SLIDE 20

20

Bell Labs

Theorem: Min Branching is NP-complete

1

r

a

r

b

r

c

r

e

r

f

r

n

r

1

C

m

C r

 

n e b a m f e c

r r r r C r r r r C , , , , , ,

1 1

 

 

n

r r r demand , , , 1  

SLIDE 21

21

Bell Labs

Conclusion:
Unicast Branching based multicast provides for efficient, policy driven Software Defined Multicast.

Nokia 2017

Policy Driven Software Defined Multicast Using Efficient Selection

f Unicast Branching Points