A Survey on Physical Network Topology Estimation


1
A Survey on Physical Network Topology Estimation

• October 21 2005
• ChikayamaTaura Lab.
• Tatsuya Shirai

2
Background

• Progress of parallel processing technologies
• Costs of parallel processing
• Cost of computation
• Cost of communication
• Clusters Grid Environments
• Cost of communication becomes bigger with larger
  scale

3
Allocation Policy

• Needs to closely allocate hosts frequently
  communicating with each other
• With multiple clusters allocate within clusters
• In single cluster allocate to use the same
  switches

4
Difficulty of estimate the cost of communication

• Shared link
• Each hosts can solely communicate at 100Mbps
• But all hosts can communicate at less than 50Mbps
  at a time
• All hosts need to work together to know this
  relation

100Mbps
5
Desired Functions

• Ideally
• Present network information to users
• Configure allocation automatically
• Needs to analyze network topology

6
Other applications

• Network trouble shooting
• Discovery of bottlenecks
• Research on routing protocol
• Simplification of local network
• etc

7
Agenda

• Background
• Network Topology
• EndtoEnd Measurement
• Researches
• Conclusion

8
Agenda

• Background
• Network Topology
• EndtoEnd Measurement
• Researches
• Conclusion

9
Network Topology

• A structure of a network
• node
• host
• router
• switch hub
• link

10
IP Layer Topology

• Structure of network
• node
• host
• router
• switch hub
• Link
• Difficulty in collecting information of LAN
  structure

11
ProtocolBased Algorithms

• Protocol
• SNMP [Yuri et al 01]
• Customized Protocol [Richard et al 04] etc.
• Hardwaredependent
• Some hubs or switches doesnt support required
  protocols.
• Deterministic estimation

12
EndtoEnd Measurement

• Metric
• Packet loss rates [Bestavros et al 02]
• Delays [Coates et al 01]
• Hardware independent
• Always possible to measure topologies of hosts
  who can communicate with root
• Probabilistic

13
Classification
14
Agenda

• Background
• Network Topology
• EndtoEnd Measurement
• Researches
• Conclusion

15
EndtoEnd Measurement

• Assume topologies are Treestructured
• Only one route exists between two hosts.
• Does not be changed while measuring
• Estimate branches of routes connecting hosts

16
Estimated topology using EndtoEnd Measurement
actual topology
estimated topology
17
EndtoEnd Measurement

• Assume topologies are Treestructured
• Only one route exists between two hosts.
• Does not be changed while measuring
• Estimate branches of routes connecting hosts
• Variance in the measurements

18
Variance of measurements

• With a small variance estimation is
  deterministic
• With a large variance estimation is
  probabilistic
• Use statistics
• Search the topology that fits the most with
  measurement

19
EndtoEnd Measurement

• Assume topologies are Treestructured
• Only one route exists between two hosts.
• Does not be changed while measuring
• Estimate branches of routes connecting hosts
• Variance in the measurements
• Procedures consist of 2 steps
• 1. Measurement
• 2. Estimation

20
Agenda

• Background
• Network Topology
• EndtoEnd Measurement
• Researches
• Conclusion

21
Researches

• Maximum Likelihood Network Topology
  Identification from edgebased unicast
  measurements [Coates et al. 01 SIGMETRICS]
• Metric Delay
• Estimation Maximum Likelihood Estimation

22
Measurement Sandwich Probe
1

• Measure delay of a link shared 2 hosts e.g. 2
  and 4
• 1. Send a small packet to 4
• 2. After constant time send a large packet to 2
• 3. Without break send a small packet to 4 again

d
3
4
2
d?d
23
Measurement
1

• The arrival of the second packet is delayed
  because the large packet is slower
• Assume that all branched nodes are not store
  forward
• Can measure delay or bandwidth of shared link
• X42 µ1d
• X32 µ1µ2d

d
µ1
µ2
3
4
2
X42
24
Estimation

• Assume delay of each shared link obeys Gaussian
  fx
• Search the topology best fitting the measurements
• ? Maximum Likelihood Estimation MLE

25
Likelihood

• The value of fitting
• Set particular topology and delay as a parameter
• Likelihood ? fXij

µ1
X32 µ1µ2
X42 µ1
µ1
µ1
26
Search Space of MLE

• Give many possible topologies to search for MLE
• Too wide to compute all topologies
• Premise
• Similar topologies have similar likelihoods
• ? Markov Chain Monte Carlo MCMC
• e.g. Hill Climbing

27
Similar Topologies Step

• Birth step
• Insert a node

• Death step
• Delete a node

1
1
1
1
2
3
4
2
3
4
2
3
4
2
3
4
28
Procedure of MLE

• 1. Give a topology at random
• 2. Make a small modification
• 3. If the new topology has greater likelihood
  adopt new topology
• 4. If a likelihood is at local maximum
• return to procedure 1
• 5. Otherwise goto2
• Can get a great likelihood topology in feasible
  time

29
Experiment

• Experimental Setup
• The root host and ten other hosts
• Measurement
• Sent 8600 probes On
• For 8 minutes
• MLE
• For 30120 seconds

2
30
The estimated topology using traceroute
The estimated topology using Coates method
31
Agenda

• Background
• Network Topology
• EndtoEnd Measurement
• Researches
• Conclusion

32
Conclusion

• Conclusion
• I Indicated importance of topology estimation and
  introduced one methods with EndtoEnd
  measurement
• Future Works
• Topology Estimation within LAN of many nodes

33
Information of Network

• Network Tomography
• Topology construction of Network
• Bandwidth Packet loss rate Shared link
• Analysis of the topology is the first step to
  estimating characteristics of a network

34
Researches

• Maximum Likelihood Network Topology
  Identification from edgebased unicast
  measurements [Coates et al. 01 SIGMETRICS]
• Metric Delay
• Estimation Maximum Likelihood Estimation
• Multicast Topology Inference from EndtoEnd
  Measurements [Duffield et al. 02 IEEE Trans]
• Metric Packet loss
• Estimation Maximum Likelihood Estimation

35
Coates Method

• Policy
• Measure delay of links shared by each pair of
  hosts
• Compute likelihood of many topologies and link
  delay
• Pick up the topology with the maximum likelihood

36
??? ????
37
Sandwitch Probing
38
Shared Link
39
Similar Topologies Step

• Methods to update topologies
• Birth step
• Death step
• µstep

µa
µe
µc
µd
40
Death step

• Delete a node reverse of Birth step

41
µstep

• Change a delay value of a link

µa
µa
µb
µb
42
Birth step

• Insert a node

43
Death step

• Delete a node reverse of Birth step

44
µstep

• Change a delay value of a link

µa
µa
µb
µb
45
d
d?d
46
No Transcript
47
The actual topology
The estimated topology