predicting internet network distance with coordinates-based
TRANSCRIPT
![Page 1: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/1.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 1
Predicting Internet Network Distance with Coordinates-Based Approaches
T.S. Eugene Ng and Hui ZhangDepartment of Computer Science
Carnegie Mellon University
![Page 2: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/2.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 2
Problem Statement
• Network distance– Round-trip propagation and transmission delay– Relatively stable, may be predictable
• Given two Internet hosts, can we accurately estimate the network distance between them without sending any RTT probes between them?
![Page 3: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/3.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 3
Why Predict Network Distance?
• Want to measure network performance to improve performance of applications– Napster, content addressable overlays, overlay multicast
• Huge number of paths to measure• TCP bandwidth and RTT probes are time-consuming
• Predicted network distance enables fast and scalable first-order performance optimization– Eliminate poor choices– Refine when needed
![Page 4: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/4.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 4
IDMaps [Francis et al. ‘99]
• Servers maintain simplified topological map
A
B
![Page 5: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/5.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 4
IDMaps [Francis et al. ‘99]
• Servers maintain simplified topological map
A
B
Tracer
Tracer
Tracer
![Page 6: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/6.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 4
IDMaps [Francis et al. ‘99]
• Servers maintain simplified topological map
A
B
Tracer
Tracer
Tracer
![Page 7: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/7.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 4
IDMaps [Francis et al. ‘99]
• Servers maintain simplified topological map
A
B
Tracer
Tracer
Tracer
![Page 8: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/8.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 4
IDMaps [Francis et al. ‘99]
• Servers maintain simplified topological map
A
B
Tracer
Tracer
Tracer
Server
![Page 9: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/9.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 4
IDMaps [Francis et al. ‘99]
• Servers maintain simplified topological map
A
B
Tracer
Tracer
Tracer
Server
A/B
![Page 10: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/10.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 4
IDMaps [Francis et al. ‘99]
• Servers maintain simplified topological map
A
B
Tracer
Tracer
Tracer
Server
![Page 11: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/11.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 4
IDMaps [Francis et al. ‘99]
• Servers maintain simplified topological map
A
B
Tracer
Tracer
Tracer
Server50ms
![Page 12: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/12.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 5
Disadvantage of Client-Server Architecture
• Unavoidable additional delay in communicating with servers
• Shared servers can become performance bottleneck
![Page 13: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/13.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 6
Can this Problem be Solved with a Peer-to-Peer Architecture?
• Flip the problem: Can end hosts maintain “coordinates” that describe their network locations?
• End hosts exchange coordinates to compute distance– High performance, high scalability
(10,50,22) (-9,-10,3)
Distance function
68ms
![Page 14: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/14.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 7
Approach 1: Global Network Positioning (GNP) Coordinates
• Model the Internet as a geometric space (e.g. 3-D Euclidean)
• Characterize the position of any end host with geometric coordinates
• Use geometric distances to predict network distances
y(x2,y2,z2)
x
z
(x1,y1,z1)
(x3,y3,z3)(x4,y4,z4)
![Page 16: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/16.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 8
GNP Landmark Operationsy
xInternet
L1
L2
L3
• Small number of distributed hosts called Landmarks measure inter-Landmark distances
![Page 17: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/17.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 8
GNP Landmark Operationsy
xInternet
L1
L2
L3
• Small number of distributed hosts called Landmarks measure inter-Landmark distances
Internet
L1
L
L
![Page 18: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/18.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 8
GNP Landmark Operationsy
xInternet
L1
L2
L3
• Small number of distributed hosts called Landmarks measure inter-Landmark distances
Internet
L1
L
L
• Compute Landmark coordinates by minimizing the overall discrepancy between measured distancesand computed distances– Cast as a generic multi-dimensional global minimization
problem
![Page 19: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/19.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 8
GNP Landmark Operationsy
xInternet
L1
L2
L3
• Small number of distributed hosts called Landmarks measure inter-Landmark distances
Internet
L1
L
L
• Compute Landmark coordinates by minimizing the overall discrepancy between measured distancesand computed distances– Cast as a generic multi-dimensional global minimization
problem
(x2,y2)
(x1,y1)
(x3,y3)
L1
L2
L3
![Page 20: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/20.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 9
GNP Ordinary Host Operations
x
Internet
(x2,y2)
(x1,y1)
(x3,y3)
L1
L2
L3
yL2
L1
L3
![Page 21: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/21.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 9
GNP Ordinary Host Operations
x
Internet
(x2,y2)
(x1,y1)
(x3,y3)
L1
L2
L3
yL2
L1
L3
• Each ordinary host measures its distances to the Landmarks, Landmarks just reflect pings
![Page 22: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/22.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 9
GNP Ordinary Host Operations
x
Internet
(x2,y2)
(x1,y1)
(x3,y3)
L1
L2
L3
yL2
L1
L3
• Each ordinary host measures its distances to the Landmarks, Landmarks just reflect pings
Internet
L1
L2
L3
![Page 23: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/23.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 9
GNP Ordinary Host Operations
x
Internet
(x2,y2)
(x1,y1)
(x3,y3)
L1
L2
L3
yL2
L1
L3
• Each ordinary host measures its distances to the Landmarks, Landmarks just reflect pings
Internet
L1
L2
L3
• Ordinary host computes its own coordinates relative to the Landmarks by minimizing the overall discrepancy between measured distances and computed distances– Cast as a generic multi-dimensional global minimization
problem
![Page 24: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/24.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 9
GNP Ordinary Host Operations
x
Internet
(x2,y2)
(x1,y1)
(x3,y3)
L1
L2
L3
yL2
L1
L3
• Each ordinary host measures its distances to the Landmarks, Landmarks just reflect pings
Internet
L1
L2
L3
• Ordinary host computes its own coordinates relative to the Landmarks by minimizing the overall discrepancy between measured distances and computed distances– Cast as a generic multi-dimensional global minimization
problem
(x4,y4)
L2
L1
L3
![Page 25: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/25.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 10
Important Questions
• What geometric model to use? • How to measure error in minimizations?• How to select Landmarks?• How many Landmarks?• What are the sources of prediction error?• How to reduce overhead?• Can we use geographical coordinates?
• Please see our paper• This talk: focus on performance comparisons
![Page 26: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/26.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 11
Approach 2: Triangulated Heuristic Coordinates
• Proposed by Hotz in 1994 for A* heuristic shortest network path search
• Provides upper and lower bounds for network distance– Assumes shortest path routing enforced
• This paper is the first study to apply and evaluate triangulated heuristic as a network distance prediction mechanism
![Page 27: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/27.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 12
Approach 2: Triangulated Heuristic Coordinates
BA
Hop count is used in this example
![Page 28: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/28.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 12
Approach 2: Triangulated Heuristic Coordinates
BA
Hop count is used in this example
Base2
Base1
![Page 29: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/29.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 12
Approach 2: Triangulated Heuristic Coordinates
BA
Hop count is used in this example
Base2
Base1( , )
( , )
a1 a2b1 b2
Base1
![Page 30: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/30.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 12
Approach 2: Triangulated Heuristic Coordinates
BA
Hop count is used in this example
Base2
Base1( , )
( , )
a1 a2b1 b2
Base1
![Page 31: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/31.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 12
Approach 2: Triangulated Heuristic Coordinates
BA
Hop count is used in this example
Base2
Base1( , )
( , )
a1 a2b1 b2
Base11(
1
![Page 32: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/32.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 12
Approach 2: Triangulated Heuristic Coordinates
BA
Hop count is used in this example
Base2
Base1( , )
( , )
a1 a2b1 b2
Base11(
1
A
1
![Page 33: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/33.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 12
Approach 2: Triangulated Heuristic Coordinates
BA
Hop count is used in this example
Base2
Base1( , )
( , )
a1 a2b1 b2
Base11(
1
2, 2
![Page 34: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/34.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 12
Approach 2: Triangulated Heuristic Coordinates
BA
Hop count is used in this example
Base2
Base1( , )
( , )
a1 a2b1 b2
Base11(
1
2, 2
2 ,1
![Page 35: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/35.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 12
Approach 2: Triangulated Heuristic Coordinates
BA
Hop count is used in this example
Base2
Base1( , )
( , )
a1 a2b1 b2
Base11(
1
2, 2
2 ,1
2, )2
![Page 36: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/36.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 12
Approach 2: Triangulated Heuristic Coordinates
BA
Hop count is used in this example
Base2
Base1( , )
( , )
a1 a2b1 b2
Base11(
1
2, 2
2 ,1
2, )2
Upper bound (U) = min (ai + bi)
![Page 37: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/37.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 12
Approach 2: Triangulated Heuristic Coordinates
BA
Hop count is used in this example
Base2
Base1( , )
( , )
a1 a2b1 b2
Base11(
1
2, 2
2 ,1
2, )2
Upper bound (U) = min (ai + bi) = 3
![Page 38: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/38.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 12
Approach 2: Triangulated Heuristic Coordinates
BA
Hop count is used in this example
Base2
Base1( , )
( , )
a1 a2b1 b2
Base11(
1
2, 2
2 ,1
2, )2
Upper bound (U) = min (ai + bi) = 3 Correct!
![Page 39: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/39.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 13
Evaluation Methodology
• 19 Probes– 12 in North America, 5 in East Asia, 2 in Europe
• 869 IP addresses called Targets we do not control– Span 44 countries
• Probes measure– Inter-Probe distances– Probe-to-Target distances
• See paper for more results
![Page 40: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/40.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 14
Evaluation Methodology (Cont’d)
• Choose a subset of well-distributed Probes to be Tracers/Base nodes/Landmarks
• Use the rest for evaluation
T
T
T
T
T
P3
P1
P4
P2
T
![Page 41: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/41.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 14
Evaluation Methodology (Cont’d)
• Choose a subset of well-distributed Probes to be Tracers/Base nodes/Landmarks
• Use the rest for evaluation
T
T
T
T
T
P3
P1
P4
P2
TP3
P
P2
![Page 42: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/42.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 14
Evaluation Methodology (Cont’d)
• Choose a subset of well-distributed Probes to be Tracers/Base nodes/Landmarks
• Use the rest for evaluation
T
T
T
T
T
P3
P1
P4
P2
TP3
P
P2(x1,y1)
![Page 43: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/43.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 14
Evaluation Methodology (Cont’d)
• Choose a subset of well-distributed Probes to be Tracers/Base nodes/Landmarks
• Use the rest for evaluation
T
T
T
T
T
P3
P1
P4
P2
TP3
P
P2(x1,y1)
3
P1
P
T
![Page 44: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/44.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 14
Evaluation Methodology (Cont’d)
• Choose a subset of well-distributed Probes to be Tracers/Base nodes/Landmarks
• Use the rest for evaluation
T
T
T
T
T
P3
P1
P4
P2
TP3
P
P2(x1,y1)
3
P1
P
T (x2, y2)
![Page 45: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/45.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 14
Evaluation Methodology (Cont’d)
• Choose a subset of well-distributed Probes to be Tracers/Base nodes/Landmarks
• Use the rest for evaluation
T
T
T
T
T
P3
P1
P4
P2
TP3
P
P2(x1,y1)
3
P1
P
T (x2, y2)
2
T
![Page 46: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/46.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 15
Performance Metrics
• Directional relative error– Symmetrically measure over and under predictions
• Relative error = abs(Directional relative error)
),min( predictedmeasured
measuredpredicted−
![Page 47: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/47.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 16
Relative Error Comparison
90% = 0.97
![Page 48: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/48.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 17
Relative Error Comparison
90% = 0.9790% = 0.59
![Page 49: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/49.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 18
Relative Error Comparison
90% = 0.9790% = 0.59
90% = 0.50
![Page 50: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/50.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 19
Directional Relative Error Analysis
DRE
Frequency
95%
75%
Mean50%
25%
5%
*
![Page 51: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/51.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 20
Directional Relative Error Comparison
![Page 52: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/52.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 21
Directional Relative Error Comparison
![Page 53: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/53.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 22
Directional Relative Error Comparison
![Page 54: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/54.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 23
Why the Difference in Over-Predictions?
IDMaps
Straight-line distance used in this example
![Page 55: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/55.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 23
Why the Difference in Over-Predictions?
IDMaps
Straight-line distance used in this example
![Page 56: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/56.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 23
Why the Difference in Over-Predictions?
IDMaps
Straight-line distance used in this example
Triangulated Heuristic
![Page 57: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/57.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 23
Why the Difference in Over-Predictions?
IDMaps
Straight-line distance used in this example
Triangulated Heuristic
![Page 58: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/58.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 23
Why the Difference in Over-Predictions?
IDMaps
Straight-line distance used in this example
Triangulated Heuristic
GNP (1-dimensional model)
![Page 59: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/59.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 23
Why the Difference in Over-Predictions?
IDMaps
Straight-line distance used in this example
Triangulated Heuristic
GNP (1-dimensional model)
![Page 60: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/60.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 24
Sensitivity to Infrastructure Node Placement
• Which nodes are used as Tracers/Bases/Landmarks matter
• High sensitivity means the approach is less robust• Test sensitivity by picking 20 random combinations of
6 infrastructure nodes and observe performance variance
![Page 61: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/61.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 25
6 Random Infrastructure Nodes(20 Experiments)
00.20.40.60.8
11.21.41.61.8
2
GNP TriangulatedHeuristic/U
IDMaps
90%
Rel
ativ
e E
rror
Max Mean Min Standard Deviation
![Page 62: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/62.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 26
Conclusions
• Coordinates-based approaches represent a new class of solutions
• These solutions fit well with the peer-to-peer architecture– Potentially better performance and scalability than the client-
server architecture
• Careful Internet evaluation shows that coordinates-based approaches are more accurate than IDMaps
• GNP is the most accurate and robust solution
![Page 63: Predicting Internet Network Distance with Coordinates-Based](https://reader036.vdocuments.net/reader036/viewer/2022062504/589c3fac1a28abf9498b58c0/html5/thumbnails/63.jpg)
T. S. Eugene Ng [email protected] Carnegie Mellon University 27
Internet as an Euclidean Space? Remarkable!
Internet map as of 1998 byBill Cheswick, Bell LabsHal Burch, CMU