selfish dtn capacity experimental estimation
DESCRIPTION
Selfish DTN capacity Experimental estimation. Dagstuhl Seminar DTN II February 2009. Joint work of PU Tournoux, V. Conan, J. Crowcroft, J. Leguay, F. Benbadis, M De Amorim. DTN performance. Measured in terms of packet delivery ratio, packet delay for non-congested networks - PowerPoint PPT PresentationTRANSCRIPT
![Page 1: Selfish DTN capacity Experimental estimation](https://reader035.vdocuments.site/reader035/viewer/2022062501/5681684a550346895dde3874/html5/thumbnails/1.jpg)
1
Selfish DTN capacityExperimental estimation
Joint work ofPU Tournoux, V. Conan, J. Crowcroft, J. Leguay, F. Benbadis, M De Amorim
Dagstuhl Seminar DTN IIFebruary 2009
![Page 2: Selfish DTN capacity Experimental estimation](https://reader035.vdocuments.site/reader035/viewer/2022062501/5681684a550346895dde3874/html5/thumbnails/2.jpg)
2
DTN performance
• Measured in terms of packet delivery ratio, packet delay for non-congested networks
• Performance is limited by:– Storage constraints– Transmission constraints– Node selfishness
![Page 3: Selfish DTN capacity Experimental estimation](https://reader035.vdocuments.site/reader035/viewer/2022062501/5681684a550346895dde3874/html5/thumbnails/3.jpg)
3
The problem
For a given traffic pattern, what is the maximum total demand that can be achieved
before the network becomes saturated, either in link capacity or in storage capacity?
![Page 4: Selfish DTN capacity Experimental estimation](https://reader035.vdocuments.site/reader035/viewer/2022062501/5681684a550346895dde3874/html5/thumbnails/4.jpg)
4
Capacity estimation
• Use traces of node contacts as input• Consider fixed demands
– One demand = transfer x Kbits from node s to node d injected at time t
• Traffic is routedat Wardrop equilibrium
• Increase x for all demands• Stop when storage/trans
mission overload increasesfaster than load
![Page 5: Selfish DTN capacity Experimental estimation](https://reader035.vdocuments.site/reader035/viewer/2022062501/5681684a550346895dde3874/html5/thumbnails/5.jpg)
5
Wardrop equilibrium
• The journey times (or costs) in all routes actually used are equal and less than those which would be experienced by a single vehicle (or message) on any unused route.
• For additive and separable positive and continuous cost functions the Wardrop equilibrium is a convex minimization problem. The equilibrium is unique, and may be obtained by running the Frank-Wolfe algorithm
![Page 6: Selfish DTN capacity Experimental estimation](https://reader035.vdocuments.site/reader035/viewer/2022062501/5681684a550346895dde3874/html5/thumbnails/6.jpg)
6
Time Discretized Graph
![Page 7: Selfish DTN capacity Experimental estimation](https://reader035.vdocuments.site/reader035/viewer/2022062501/5681684a550346895dde3874/html5/thumbnails/7.jpg)
7
Example
Transmission capacities of connections
![Page 8: Selfish DTN capacity Experimental estimation](https://reader035.vdocuments.site/reader035/viewer/2022062501/5681684a550346895dde3874/html5/thumbnails/8.jpg)
8
Fastest route
• Delay is costly: the fastest route is chosen
Weight on delay = 1Weight on transmission cost = 1
![Page 9: Selfish DTN capacity Experimental estimation](https://reader035.vdocuments.site/reader035/viewer/2022062501/5681684a550346895dde3874/html5/thumbnails/9.jpg)
9
Costly transmissions
• Increasing transmission cost forces usage of slower routes
Weight on delay = 1Weight on transmission cost = 30
![Page 10: Selfish DTN capacity Experimental estimation](https://reader035.vdocuments.site/reader035/viewer/2022062501/5681684a550346895dde3874/html5/thumbnails/10.jpg)
10
More delay
• Allowing for more delay forces usage of even more slow routes
Weight on delay << 1Weight on transmission cost = 30
![Page 11: Selfish DTN capacity Experimental estimation](https://reader035.vdocuments.site/reader035/viewer/2022062501/5681684a550346895dde3874/html5/thumbnails/11.jpg)
11
Infocom data set
• Capacity increases with delay tolerance
Weight on delay
1 0.50.1
0.0001
![Page 12: Selfish DTN capacity Experimental estimation](https://reader035.vdocuments.site/reader035/viewer/2022062501/5681684a550346895dde3874/html5/thumbnails/12.jpg)
12
Conclusions
• Delay-tolerance helps increase capacity.
• Routing protocols should be multi-dimensional.
• Selfishness makes network engineering challenging.