Presentation is loading. Please wait.

Presentation is loading. Please wait.

U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science R3: Robust Replication Routing in Wireless Networks with Diverse Connectivity Characteristics.

Similar presentations


Presentation on theme: "U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science R3: Robust Replication Routing in Wireless Networks with Diverse Connectivity Characteristics."— Presentation transcript:

1 U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science R3: Robust Replication Routing in Wireless Networks with Diverse Connectivity Characteristics Xiaozheng Tie, Arun Venkataramani, Aruna Balasubramanian University of Massachusetts Amherst University of Washington

2 U NIVERSITY OF M ASSACHUSETTS A MHERST Department of Computer Science 2 Wireless routing compartmentalized Protocols designed for well-connected meshes OLSR, ETT, ETX, EDR, … Protocols designed for intermittently- connected MANETs AODV, DSDV, DSR, … Protocols designed for sparsely- connected DTNs DTLSR, RAPID, Prophet, Maxprop, EBR, Random, … Research question: Can we design a simple routing protocol that ensures robust performance across networks with diverse connectivity characteristics all the way from well-connected meshes to mostly- disconnected DTNs and everything in between? Research question: Can we design a simple routing protocol that ensures robust performance across networks with diverse connectivity characteristics all the way from well-connected meshes to mostly- disconnected DTNs and everything in between?

3 U NIVERSITY OF M ASSACHUSETTS A MHERST Department of Computer Science 3 Outline Compartmentalized design harmful Quantifying replication gain R3 design and implementation Evaluation Conclusion

4 U NIVERSITY OF M ASSACHUSETTS A MHERST Department of Computer Science 4 Fragile performance Protocols perform poorly outside target environment DTN protocols perform poorly in mesh Replication wasteful Mesh protocols perform poorly in DTNs No contemporaneous path Mesh testbed DTN testbed 2.1x Normalized delay 2.2x Normalized delay

5 U NIVERSITY OF M ASSACHUSETTS A MHERST Department of Computer Science 5 Spatial connectivity diversity DieselNet-Hybrid Vehicular DTN + Wifi Mesh 20 buses in Vehicular DTN 4 open AP WiFi mesh clusters < 100 contacts 100 – 200 contacts > 200 contacts

6 U NIVERSITY OF M ASSACHUSETTS A MHERST Department of Computer Science 6 Temporal connectivity diversity Haggle Mobile ad hoc network 8 mobile and 1 stationary imotes 9 hour trace in Intel Cambridge Lab Fraction of connected nodes

7 U NIVERSITY OF M ASSACHUSETTS A MHERST Department of Computer Science 7 Compartmentalized design harmful Fragile performance under spatio- temporal diversity Makes interconnection of diverse networks difficult Manageability Separation of concerns Long-term innovation

8 U NIVERSITY OF M ASSACHUSETTS A MHERST Department of Computer Science 8 Outline Compartmentalized design harmful Quantifying replication gain R3 design and implementation Evaluation Conclusion

9 U NIVERSITY OF M ASSACHUSETTS A MHERST Department of Computer Science 9 Replication: Key difference DTN Mesh MANET Sparsely connected Well connected Intermittently connected Replication Forwarding Key question: Under what conditions and by how much replication improves performance?

10 U NIVERSITY OF M ASSACHUSETTS A MHERST Department of Computer Science 10 Model to quantify replication gain 10 Replication gain Src Dst Delay of forwarding Delay of replication Random variable denoting the delay of path i

11 U NIVERSITY OF M ASSACHUSETTS A MHERST Department of Computer Science 11 Example of replication gain 11 Replication gain Src Dst Delay of forwarding Delay of replication Replication gain depends on path delay distributions, not just expected value 5x delay improvement

12 U NIVERSITY OF M ASSACHUSETTS A MHERST Department of Computer Science 12 Trace-driven analysis on DieselNet-DTN and Haggle Replication gain vs. number of paths Vehicular DTN in DieselNet Haggle Two paths suffice to capture much of the gain

13 U NIVERSITY OF M ASSACHUSETTS A MHERST Department of Computer Science 13 Compartmentalized design harmful Quantifying replication gain R3 design and implementation Evaluation Conclusion Outline

14 U NIVERSITY OF M ASSACHUSETTS A MHERST Department of Computer Science 14 R3 design overview Link-state Estimate per-link delay distribution Replication Select replication paths using model Adapt replication to be load-aware Source routing along selected path(s) Src Dst

15 U NIVERSITY OF M ASSACHUSETTS A MHERST Department of Computer Science 15 Link delay Estimate link delay using periodic probes Estimate link delay distribution Time 30s 20s 10s 0.1s Link availability delay Delay to successfully transfer pkt Delay samples = {30.1s, 20.1s, 10.1s} Acked probe Half of round-trip delay Unacked probe Half of time since sending probe and receiving an ack for subsequent probe

16 U NIVERSITY OF M ASSACHUSETTS A MHERST Department of Computer Science 16 R3 design overview Src Dst Link-state Estimate per-link delay distribution Replication Select replication paths using model Adapt replication to be load-aware Source routing along selected path(s)

17 U NIVERSITY OF M ASSACHUSETTS A MHERST Department of Computer Science 17 First path Path s.t. it minimizes Selected using Dijkstras shortest path algorithm Second path Path s.t. it minimizes Selected using delay distributions and model Path selection using model Src Dst

18 U NIVERSITY OF M ASSACHUSETTS A MHERST Department of Computer Science 18 Problem Replication hurts performance under high load Solution Load aware replication Adapting replication to load Forwarding Replication Start actual_delay > 2 * model_estimated_delay actual_delay 2 * model_estimated_delay

19 U NIVERSITY OF M ASSACHUSETTS A MHERST Department of Computer Science 19 R3 design overview Link-state Estimate per-link delay distribution Replication Select replication paths using model Adapt replication to be load-aware Source routing along selected path(s) Src Dst

20 U NIVERSITY OF M ASSACHUSETTS A MHERST Department of Computer Science 20 Compartmentalized design harmful Quantifying replication gain R3 design and implementation Evaluation Deployment on a DTN and mesh testbed Simulation based on real traces Emulation using mesh testbed Conclusion Outline

21 U NIVERSITY OF M ASSACHUSETTS A MHERST Department of Computer Science 21 DieselNet DTN testbed 20 buses in a 150 sq. mile area Mesh testbed 16 nodes in one floor Simulator validation using DieselNet deployment < 10% of deployment result R3 Deployment

22 U NIVERSITY OF M ASSACHUSETTS A MHERST Department of Computer Science 22 Experimental settings Temporal diversity inherent in Haggle Spatial diversity inherent in DieselNet-Hybrid Varying load Compared protocols Replication: RAPID, Random Forwarding: DTLSR, AODV, OLSR Multi-configuration: SWITCH (RAPID+OLSR) R3 Trace-driven simulation

23 U NIVERSITY OF M ASSACHUSETTS A MHERST Department of Computer Science Robustness to spatial diversity Simulation based on DieselNet-Hybrid trace 23 R3 improves median delay by 2.1x 1 2 34 5 6 7 8 9

24 U NIVERSITY OF M ASSACHUSETTS A MHERST Department of Computer Science Robustness to varying load Simulation based on DieselNet-Hybrid trace 24 R3 reduces delay by up to 2.2x over SWITCH

25 U NIVERSITY OF M ASSACHUSETTS A MHERST Department of Computer Science 25 Compartmentalized design harmful R3 ensures robust performance across diverse connectivity characteristics Unified link metric based on delay distributions Replication based on delay uncertainty model Adaptive replication based on network load Conclusion

26 U NIVERSITY OF M ASSACHUSETTS A MHERST Department of Computer Science Q&A 26

27 U NIVERSITY OF M ASSACHUSETTS A MHERST Department of Computer Science 27 Theorem: Replication gain is high iff path delays are highly unpredictable Predictability of a random variable X = Smallest such that When is replication gain high? Corollary: Replication can yield unbounded gain even with two paths

28 U NIVERSITY OF M ASSACHUSETTS A MHERST Department of Computer Science 28 Path delay Expected delay: Delay distribution of : convolutions of Estimating path delay distribution Link delay distribution Path delay distribution

29 U NIVERSITY OF M ASSACHUSETTS A MHERST Department of Computer Science Robustness to temporal diversity Simulation based on Haggle trace 29 R3 reduces delay by up to 60% R3 increases goodput by up to 30%

30 U NIVERSITY OF M ASSACHUSETTS A MHERST Department of Computer Science Emulating intermediate connectivity Mesh-based emulation approach Brings link up and down to vary connectivity Emulates connectivity diversity (but not mobility) 30 R3 reduces delay by up to 2.2x


Download ppt "U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science R3: Robust Replication Routing in Wireless Networks with Diverse Connectivity Characteristics."

Similar presentations


Ads by Google