Supporting Cooperative Caching in Disruption Tolerant Networks
description
Transcript of Supporting Cooperative Caching in Disruption Tolerant Networks
![Page 1: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/1.jpg)
Supporting Cooperative Caching in Disruption Tolerant Networks
Wei Gao and Guohong CaoDept. of Computer Science and EngineeringPennsylvania State University
Arun Iyengar and Mudhakar SrivatsaIBM T. J. Watson Research Center
![Page 2: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/2.jpg)
Outline
IntroductionNetwork Central Locations (NCLs)Caching SchemePerformance EvaluationSummary & Future Work
![Page 3: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/3.jpg)
Disruption Tolerant Networks (DTNs)Consist of hand-held personal mobile devices
Laptops, PDAs, SmartphonesOpportunistic and intermittent network
connectivityResult of node mobility, device power outage or
malicious attacksHard to maintain end-to-end communication links
Data transmission via opportunistic contacts
![Page 4: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/4.jpg)
Data Transmission in DTNs Carry-and-Forward
Mobile nodes physically carry data as relaysForwarding data opportunistically upon contactsMajor problem: appropriate relay selection
B
A C
0.7
0.5
![Page 5: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/5.jpg)
Providing Data Access in DTNsActive data dissemination
Data source actively push data to users being interested in the data
Publish/SubscribeBrokers forward data to users according to their
subscriptionsCaching
![Page 6: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/6.jpg)
Our FocusCooperative caching in DTNs
Cache data at appropriate network locationsQueries in the future are responded faster
Major challengesWhere to cache
Hard to monitor the network query pattern and determine the caching location for minimizing data access delay
How many to cache Uncertainty of opportunistic data transmission Tradeoff between data accessibility and caching overhead
![Page 7: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/7.jpg)
ChallengesWireless ad-hoc network with end-to-end network
connectivityA
FB
C
D
E
H
K
q1(d1)d1
d2
q2(d1)
q3(d2)
G
d1
d2
Incidentally cache pass-by data
![Page 8: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/8.jpg)
A
FB
C
D
E
H
K
q1(d1)d1
d2
q2(d1)
q3(d2)
G
ChallengesDisruption-Tolerant Networks (DTNs)
d1
d2
![Page 9: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/9.jpg)
Basic Ideas Intentionally cache data at a set of Network
Central Locations (NCLs)A group of mobile nodes represented by a central nodeBeing easily accessed by other nodes in the network
Utility-based cache replacementPopular data is cached near central nodes
Coordinate multiple caching nodesTradeoff between data accessibility and caching
overhead
![Page 10: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/10.jpg)
The Big Picture
DataQuery
A
C1C2
B
C
D
S
RNCL 1
NCL 2
![Page 11: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/11.jpg)
Outline
IntroductionNetwork Central Locations (NCLs)Caching SchemePerformance EvaluationSummary & Future Work
![Page 12: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/12.jpg)
Network ModelingContacts in DTNs are described by network
contact graph G(V, E)Contact process between nodes i and j is modeled as
an edge eij on network contact graphThe contacts between nodes i and j are modeled as a
Poisson process with contact rate
![Page 13: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/13.jpg)
NCL SelectionCentral nodes representing NCLs are selected
based on a probabilistic metricCentral nodes can be easily accessed by other nodesNumber (K) of NCLs is a pre-defined parameter
Central nodes are selected in a centralized manner based on global network knowledge
![Page 14: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/14.jpg)
NCL Selection MetricMetric of node i to be selected as central node
Opportunistic path:
Average probability that data can be transmitted from a random non-central node to i within time T
Set of existing central nodes
![Page 15: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/15.jpg)
Trace-based ValidationThe applicability of NCL selection in realistic
DTN tracesOnly few nodes have high metric valuesOur metric reflects the heterogeneity of node contact
pattern
![Page 16: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/16.jpg)
Outline
IntroductionNetwork Central Locations (NCLs)Caching SchemePerformance EvaluationSummary & Future Work
![Page 17: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/17.jpg)
Caching Scheme Overview Initial caching locations
Data source pushes data to NCLsQuerying data
Requesters pulls data from NCLs by multicasting queries to NCLs
Utility-based cache replacementWhen caching nodes contact each otherMore popular data is cached nearer to central nodes
![Page 18: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/18.jpg)
Initial Caching LocationsCentral nodes are prioritized to cache dataData is cached at nodes near central nodes if their
buffer are full
Cache
S
C1
C3
C2
R11R11
R12R12
R13R13
R21R21 R2
2R22 R2
3R23
R31R31 R3
2R32 R3
3R33
R24R24
R34R34
![Page 19: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/19.jpg)
Querying DataRequester multicasts query to central nodes
If data is cached at central nodes Respond directly
Otherwise Forward query
to caching nodesR
C1
C3
C2A
B
![Page 20: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/20.jpg)
Analyzing Data Access DelayRelated to the number (K) of NCLsK is small
1./3.: Data transmission delay between requester and NCLs is longer
2.: Data can be cached nearer to the central node
K is largeMetric values of some central
nodes may not be high1.
2.
3.
![Page 21: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/21.jpg)
Utility-based Cache ReplacementTwo caching nodes exchange their cached data
when they contact each otherCache popular data near central nodes to minimize the
cumulative data access delayData utility
The popularity of data in the networkThe distance from the caching node to the
corresponding central node
![Page 22: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/22.jpg)
Utility-based Cache ReplacementKnapsack formulation
Data at nodes A and B:
Data utilities
Data sizes
Each data is only cached at one node
![Page 23: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/23.jpg)
Utility-based Cache ReplacementExample: A is nearer to the central node
![Page 24: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/24.jpg)
Utility-based Cache ReplacementLess popular data may be removed from cache if
caching buffer is very limited
![Page 25: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/25.jpg)
Outline
IntroductionNetwork Central Locations (NCLs)Caching SchemePerformance EvaluationSummary & Future Work
![Page 26: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/26.jpg)
Experimental SettingsData generation
Each node periodically determines whether to generate new data with a fixed probability pG=0.2
Data lifetime is uniformly distributed in [0.5T, 1.5T]Query pattern
Randomly generated at all nodes periodicallyQuery pattern follows Zipf distributionTime constraint of query: T/2
![Page 27: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/27.jpg)
Caching PerformanceOn the MIT Reality trace with different data
lifetime (T)
Lower caching overheadHigher successful ratio
![Page 28: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/28.jpg)
Effectiveness of Cache ReplacementDifferent replacement strategies on MIT Reality
trace
Improves successful ratio Similar replacement overhead
![Page 29: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/29.jpg)
Impact of the Number of NCLsOn Infocom06 trace with T=3 hours
Optimal: K=5
![Page 30: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/30.jpg)
SummaryCooperative caching in DTNs
Intentionally cache data at a pre-specified set of NCLs which can be easily accessed by other nodes
NCL selection based on a probabilistic metricUtility-based cache replacement to maximize the
cumulative caching performanceFuture work
Distributed NCL selectionLoad balancing on NCLs
![Page 31: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/31.jpg)
Thank you!
http://mcn.cse.psu.edu
The paper and slides are also available at:http://www.cse.psu.edu/~wxg139
![Page 32: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/32.jpg)
TracesRecord user contacts at university campusVarious wireless interfaces
Bluetooth: periodically detect nearby peersWiFi: associate to the best Access Point (AP)
![Page 33: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/33.jpg)
Opportunistic PathEach hop corresponds to stochastic contact
process with pairwise contact rates
Xk: inter-contact time between nodes Nk and Nk+1
Exponentially distributed Y: the time needed to transmit data from A to B along
the path follows hypoexponential distribution
![Page 34: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/34.jpg)
Probabilistic Data Selection in Cache Replacement
Fairness of cachingEvery node caches popular dataCaching effectiveness is only locally maximized
Probabilistic data selectionEvery caching node probabilistically determine
whether to cache the data Data utility is used as the probability
![Page 35: Supporting Cooperative Caching in Disruption Tolerant Networks](https://reader035.fdocuments.us/reader035/viewer/2022062812/56816380550346895dd461dd/html5/thumbnails/35.jpg)
ComparisonsNoCache, where caching is not used for data
accessRandomCache, where each requester caches the
received data randomlyCacheData, which is proposed for cooperative
caching in wireless ad-hoc networksBundleCache, which caches network data as
bundles