Feb 2003 2 nd IPTPS Lighthouses for Scalable Distributed Location Marcelo Pias UCL Jon Crowcroft...
-
Upload
angelina-francis -
Category
Documents
-
view
215 -
download
0
Transcript of Feb 2003 2 nd IPTPS Lighthouses for Scalable Distributed Location Marcelo Pias UCL Jon Crowcroft...
![Page 1: Feb 2003 2 nd IPTPS Lighthouses for Scalable Distributed Location Marcelo Pias UCL Jon Crowcroft CL/Cambridge University Steve Wilbur UCL Tim Harris Cambridge.](https://reader036.fdocuments.in/reader036/viewer/2022072016/56649ee15503460f94bf209a/html5/thumbnails/1.jpg)
Feb 2003 2nd IPTPS
Lighthouses for Scalable Distributed Location
Marcelo Pias UCLJon Crowcroft CL/Cambridge UniversitySteve Wilbur UCLTim Harris Cambridge UniversitySaleem Bhatti UCL
![Page 2: Feb 2003 2 nd IPTPS Lighthouses for Scalable Distributed Location Marcelo Pias UCL Jon Crowcroft CL/Cambridge University Steve Wilbur UCL Tim Harris Cambridge.](https://reader036.fdocuments.in/reader036/viewer/2022072016/56649ee15503460f94bf209a/html5/thumbnails/2.jpg)
Feb 2003 2nd IPTPS
Motivation
Self-organizing content addressable storage
Networked Games
Resources discovery (e.g. GRID)
Finesse failure of “triangle inequality”
Questions:
a. “Could we characterise network proximity in a scalable model by
computing node locations with a set of coordinates?
b. Could this model help these systems in selecting topologically close peers?”
![Page 3: Feb 2003 2 nd IPTPS Lighthouses for Scalable Distributed Location Marcelo Pias UCL Jon Crowcroft CL/Cambridge University Steve Wilbur UCL Tim Harris Cambridge.](https://reader036.fdocuments.in/reader036/viewer/2022072016/56649ee15503460f94bf209a/html5/thumbnails/3.jpg)
Feb 2003 2nd IPTPS
Problem
Constraint: only a few distance measures are available.
General Space ‘M’
Objects {x1,x2,…,xn} = Network nodes
Distance Measures (e.g. propagation delay)
x1
x2
x3
Mapping
Vector Space ‘V’
Points {v1,v2,…,vn}
Distances Preserved
v2
v1
v3
![Page 4: Feb 2003 2 nd IPTPS Lighthouses for Scalable Distributed Location Marcelo Pias UCL Jon Crowcroft CL/Cambridge University Steve Wilbur UCL Tim Harris Cambridge.](https://reader036.fdocuments.in/reader036/viewer/2022072016/56649ee15503460f94bf209a/html5/thumbnails/4.jpg)
Feb 2003 2nd IPTPS
Pivoting
Lighthouse
Multiple local bases (decentralised)
Transition Matrix
Global basis
Same well-known ‘pivot nodes’
Scalability problem (bottlenecks/node failure)
Pivoting
Distance from a given node to pre-selected pivots {p1,…,pn}
![Page 5: Feb 2003 2 nd IPTPS Lighthouses for Scalable Distributed Location Marcelo Pias UCL Jon Crowcroft CL/Cambridge University Steve Wilbur UCL Tim Harris Cambridge.](https://reader036.fdocuments.in/reader036/viewer/2022072016/56649ee15503460f94bf209a/html5/thumbnails/5.jpg)
Feb 2003 2nd IPTPS
n6
n4 n
5
n2
n1
n3
G = {n1n2, , n1n3 }
{n 4,n 5,
, n6 }
REQ Lig
hthouse
s
Distance Measure(e.g. Ping)
1. Finding Lighthouses
n7
![Page 6: Feb 2003 2 nd IPTPS Lighthouses for Scalable Distributed Location Marcelo Pias UCL Jon Crowcroft CL/Cambridge University Steve Wilbur UCL Tim Harris Cambridge.](https://reader036.fdocuments.in/reader036/viewer/2022072016/56649ee15503460f94bf209a/html5/thumbnails/6.jpg)
Feb 2003 2nd IPTPS
n6
n4 n
5
n2
n1
n3
G = {n1n2, , n1n3 }
2. Local Basis Coordinates
L = {n4n5, , n4n6}
n7
![Page 7: Feb 2003 2 nd IPTPS Lighthouses for Scalable Distributed Location Marcelo Pias UCL Jon Crowcroft CL/Cambridge University Steve Wilbur UCL Tim Harris Cambridge.](https://reader036.fdocuments.in/reader036/viewer/2022072016/56649ee15503460f94bf209a/html5/thumbnails/7.jpg)
Feb 2003 2nd IPTPS
n6
n4
n5
n2
n1
n3
G = {n1n2, , n1n3 }
3. Host Coordinates
n7 = c1 . l1 + c2 . l2
l1 = {n4n5 }
l2 = {n4n6 }
c1 . l
1
c2 . l2
![Page 8: Feb 2003 2 nd IPTPS Lighthouses for Scalable Distributed Location Marcelo Pias UCL Jon Crowcroft CL/Cambridge University Steve Wilbur UCL Tim Harris Cambridge.](https://reader036.fdocuments.in/reader036/viewer/2022072016/56649ee15503460f94bf209a/html5/thumbnails/8.jpg)
Feb 2003 2nd IPTPS
n6
n4
n5
n2
n1
n3
G = {n1n2, , n1n3 }
4. Transition Matrix
n7 = c1 . l1 + c2 . l2
l1 = {n4n5 }
l2 = {n4n6 }
n7 maintains a transition matrix
P = [ [n4n5] [n4n6] ]
![Page 9: Feb 2003 2 nd IPTPS Lighthouses for Scalable Distributed Location Marcelo Pias UCL Jon Crowcroft CL/Cambridge University Steve Wilbur UCL Tim Harris Cambridge.](https://reader036.fdocuments.in/reader036/viewer/2022072016/56649ee15503460f94bf209a/html5/thumbnails/9.jpg)
Feb 2003 2nd IPTPS
Initial Experiments
Objectives
Accuracy: how close the predicted distance is to the real distance measured
Data (delay measures available at http://www-2.cs.cmu.edu/~eugeneng/research)
Probe matrix: mutual distance measures between 19 probes around the World
Target matrix: delay measures between 869 hosts and the 19 probes
Dimensions Distance Function
Number Probes
3 L2 (Euclidean) 4
Table I: Key Parameters
![Page 10: Feb 2003 2 nd IPTPS Lighthouses for Scalable Distributed Location Marcelo Pias UCL Jon Crowcroft CL/Cambridge University Steve Wilbur UCL Tim Harris Cambridge.](https://reader036.fdocuments.in/reader036/viewer/2022072016/56649ee15503460f94bf209a/html5/thumbnails/10.jpg)
Feb 2003 2nd IPTPS
Initial Results
![Page 11: Feb 2003 2 nd IPTPS Lighthouses for Scalable Distributed Location Marcelo Pias UCL Jon Crowcroft CL/Cambridge University Steve Wilbur UCL Tim Harris Cambridge.](https://reader036.fdocuments.in/reader036/viewer/2022072016/56649ee15503460f94bf209a/html5/thumbnails/11.jpg)
Feb 2003 2nd IPTPS
Open Questions
Network Performance Metrics
Propagation delay so far. What about other metrics?
E.g. power,cost,txput… Distance Function
Metric or non-metric?
E.g. power metric might reflect attenuation
What about obstacles (can we reflect shadows)? Curse of Dimensionality
How many dimensions?
Crovella reports <= 7 works very well on large internet
What would work for other metric in ad hoc wireless Choosing Lighthouses
Do they form a linear independent basis?
How could Lighthouse be incorporated in DHT/DT systems?
![Page 12: Feb 2003 2 nd IPTPS Lighthouses for Scalable Distributed Location Marcelo Pias UCL Jon Crowcroft CL/Cambridge University Steve Wilbur UCL Tim Harris Cambridge.](https://reader036.fdocuments.in/reader036/viewer/2022072016/56649ee15503460f94bf209a/html5/thumbnails/12.jpg)
Feb 2003 2nd IPTPS
Conclusions
Lighthouse
maps objects and their distance measures onto points in a k-dimensional vector space.
It avoids the scalability problem presented by systems that rely on ‘well-known’ pivots as reference points (e.g. GNP, Binning, Beaconing)
It computes coordinates as accurate as GNP
Future work
Investigate the ‘open questions’