A SimulationBased Study of Overlay Routing Performance

1
A Simulation-Based Study of Overlay Routing
Performance

• CS 268 Course Project
• Andrey Ermolinskiy, Hovig Bayandorian, Daniel Chen

2
Problem Statement

• The current Internet infrastructure is highly
  resistant to change new network-layer
  functionality is difficult to implement and
  costly to deploy.
• Standard Internet routing protocols (BGP, OSPF)
  generally fail to deliver low- latency and
  high-throughput routing paths.
• Coarse-grained routing metrics (e.g., hop count)
• BGP policies
• Lack of economic incentives to provide
  performance-based routing
• Overlay networks offer an alternative method for
  deploying new routing functionality
• Set up an application-level logical network.
• Deploy a higher-level routing protocol that
  routes packets according to any desired routing
  metric.
• Question 1 Given an alternative routing protocol
  that implements an arbitrary routing metric, to
  what extent can an overlay-based solution
  approximate a network-layer deployment?
• Question 2 What is the best way to select
  overlay nodes?

3
Approach

• Assume the existence of two distinct routing
  metrics WB (base), WO (overlay) and two
  routing protocols RB, RO that implement
  shortest path routing for WB and WO ,
  respectively.
• Model the topology as a dualweighted graph
• G (V, WB, WO).
• RB is the default routing algorithm in G, but a
  subset of nodes O implements RO via an overlay
  mechanism.

CO(V1-V2-V5-V6) 17
CO(V1-V3-V2-V4-V5-V6) 9
4
Approach

• To evaluate the performance of a given choice of
  O, compute the topology stretch.

Logical overlay subgraph
Dual-weighted topology graph
Stretch(v1, v6) 11 / 9
O v2, v4, v5
Stretch(v1, v6) 9 / 9 1 (Optimal Path)
O v2, v3 , v4, v5
5
Results Distribution of Stretch Values

• Exhaustive search through the space of all
  subsets of nodes in two uniform random topologies
  from ?(N, p),

N 20, p 0.5
N 20, p 0.1

• Only 2 overlay nodes are required to achieve
  optimal routing.
• 90 of all overlays of size 11 or more achieve
  stretch value of 1.03.

• 11 overlay nodes are required to achieve optimal
  routing.

6
Average Stretch in Large Random Topologies

• Estimated average stretch for three uniform
  random topologies drawn from ?(1000, p).
• A large fraction of the maximum achievable
  improvement lies with small selections of overlay
  nodes

For p 0.1 100 overlay nodes provide 66.0 of
the achievable improvement. For p 0.5 100
overlay nodes provide 86.4. For p 0.9 100
overlay nodes provide 89.9.
7
Choice of the Topology Model
Uniform random from ?(1050, 0.0196).
Waxman (N 1050, a 0.086, ß 0.2).
Transit-stub 1 transit domain with 50 nodes 1
stub domain per transit node 20 nodes per stub
domain, each domain is a complete subgraph.
8
Alternative Weight Distributions

• Edge weights WB(vi, vj) and WO(vi,vj) were drawn
  from a bivariate Gaussian distribution.

Positively correlated weights
Negatively correlated weights

• Gaussian with negatively correlated weights
  produces stretch bounded away from 1.
• Distribution does not appear to affect the shape
  of the curve.

9
Computational Complexity of Optimal Overlay Search
OPTOVERLAY(G, k) Given a dual-weighted graph G,
find an optimal overlay of size at most k if one
exists.

• We can show that OPTOVERLAY is NP-complete
• The proof involves reducing 3-SAT to the decision
  problem Does there exist an optimal overlay of
  size k?

The figure to the left illustrates a sample
construction that corresponds to the Boolean
formula (x1 x2 x3) (x2 x3 x4)
10
Node Selection Heuristics

• We investigated several node selection
  heuristics, including
• Selecting a random subset
• Selecting the best of k random subsets
• Degree sorting
• selecting nodes with the highest degree
• Importance sorting
• selecting the most important nodes
• Node importance is defined as the number of
  pairwise shortest paths in the topology to which
  the node belongs.
• Structure-aware selection (for transit-stub
  topologies)
• Selecting only the transit nodes
• Selecting only the periphery nodes
• Selecting one node from each domain, etc.

11
Heuristics Results

• Transit-stub topology
• 4 transit domains
• 25 nodes per transit domain
• 1 stub domain per transit node
• 9 nodes per stub domain

• Waxman topology
• N 1050, a 0.086, ß 0.2

• Degree sorting works best.

• 1 node from every stub domain works best.

12
Summary and Conclusions

• Overlay networks work well on a range of
  simulated topology models and most of the
  improvement can be achieved with a fairly small
  selection of overlay nodes.
• Picking an optimal overlay is hard, but degree
  sorting and other heuristics provide efficient
  approximations.
• Random node selection works surprisingly well.

13
Heuristics Results