When processing is cheaper than transmitting Daniel V Uhlig Maryam Rahmaniheris 1.

Slides:



Advertisements
Similar presentations
February 20, Spatio-Temporal Bandwidth Reuse: A Centralized Scheduling Mechanism for Wireless Mesh Networks Mahbub Alam Prof. Choong Seon Hong.
Advertisements

Multicasting in Mobile Ad hoc Networks By XIE Jiawei.
Data and Computer Communications
1 S4: Small State and Small Stretch Routing for Large Wireless Sensor Networks Yun Mao 2, Feng Wang 1, Lili Qiu 1, Simon S. Lam 1, Jonathan M. Smith 2.
Multicast in Wireless Mesh Network Xuan (William) Zhang Xun Shi.
Bidding Protocols for Deploying Mobile Sensors Reporter: Po-Chung Shih Computer Science and Information Engineering Department Fu-Jen Catholic University.
Maximum Battery Life Routing to Support Ubiquitous Mobile Computing in Wireless Ad Hoc Networks By C. K. Toh.
Decentralized Reactive Clustering in Sensor Networks Yingyue Xu April 26, 2015.
TDMA Scheduling in Wireless Sensor Networks
CLUSTERING IN WIRELESS SENSOR NETWORKS B Y K ALYAN S ASIDHAR.
Highly-Resilient, Energy-Efficient Multipath Routing in Wireless Sensor Networks Computer Science Department, UCLA International Computer Science Institute,
1 Routing Techniques in Wireless Sensor networks: A Survey.
Search and Replication in Unstructured Peer-to-Peer Networks Pei Cao, Christine Lv., Edith Cohen, Kai Li and Scott Shenker ICS 2002.
Monday, June 01, 2015 ARRIVE: Algorithm for Robust Routing in Volatile Environments 1 NEST Retreat, Lake Tahoe, June
SYNOPSIS DIFFUSION For Robust Aggregation in Sensor Networks Suman Nath, Phillip B. Gibbons, Srinivasan Seshan, Zachary R. Anderson Presented by Xander.
Directed Diffusion: A Scalable and Robust Communication Paradigm for Sensor Networks.
Volcano Routing Scheme Routing in a Highly Dynamic Environment Yashar Ganjali Stanford University Joint work with: Nick McKeown SECON 2005, Santa Clara,
DNA Research Group 1 CountTorrent: Ubiquitous Access to Query Aggregates in Dynamic and Mobile Sensor Networks Abhinav Kamra, Vishal Misra and Dan Rubenstein.
More routing protocols Alec Woo June 18 th, 2002.
Tributaries and Deltas: Efficient and Robust Aggregation in Sensor Network Streams Amit Manjhi, Suman Nath, Phillip B. Gibbons Carnegie Mellon University.
Aggregation in Sensor Networks NEST Weekly Meeting Sam Madden Rob Szewczyk 10/4/01.
CPSC 689: Discrete Algorithms for Mobile and Wireless Systems
Probabilistic Data Aggregation Ling Huang, Ben Zhao, Anthony Joseph Sahara Retreat January, 2004.
Taming the Underlying Challenges of Reliable Multihop Routing in Sensor Networks.
LPT for Data Aggregation in Wireless Sensor networks Marc Lee and Vincent W.S Wong Department of Electrical and Computer Engineering, University of British.
Online Data Gathering for Maximizing Network Lifetime in Sensor Networks IEEE transactions on Mobile Computing Weifa Liang, YuZhen Liu.
Rethinking Internet Traffic Management: From Multiple Decompositions to a Practical Protocol Jiayue He Princeton University Joint work with Martin Suchara,
1-1 CMPE 259 Sensor Networks Katia Obraczka Winter 2005 Routing Protocols II.
Comb, Needle, and Haystacks: Balancing Push and Pull for Information Discovery Xin Liu Department of Computer Science University of California, Davis Joint.
Xingbo Yu ()ICS280sensors Winter 2005 Tributaries and Deltas: Efficient and Robust Aggregation in Sensor Networks A.ManJhi, S. Nath P. Gibbons CMU.
Beacon Vector Routing: Scalable Point-to-Point Routing in Wireless Sensornets.
CS 580S Sensor Networks and Systems Professor Kyoung Don Kang Lecture 7 February 13, 2006.
Connected Dominating Sets in Wireless Networks My T. Thai Dept of Comp & Info Sci & Engineering University of Florida June 20, 2006.
Delay Efficient Sleep Scheduling in Wireless Sensor Networks Gang Lu, Narayanan Sadagopan, Bhaskar Krishnamachari, Anish Goel Presented by Boangoat(Bea)
T AG : A TINY AGGREGATION SERVICE FOR AD - HOC SENSOR NETWORKS Samuel Madden, Michael J. Franklin, Joseph Hellerstein, and Wei Hong Presented by – Mahanth.
TAG: a Tiny Aggregation Service for Ad-Hoc Sensor Networks Paper By : Samuel Madden, Michael J. Franklin, Joseph Hellerstein, and Wei Hong Instructor :
Information Quality Aware Routing in Event-Driven Sensor Networks Hwee-Xian TAN 1, Mun Choon CHAN 1, Wendong XIAO 2, Peng-Yong KONG 2 and Chen-Khong THAM.
CS2510 Fault Tolerance and Privacy in Wireless Sensor Networks partially based on presentation by Sameh Gobriel.
Decentralized Scattering of Wake-up Times in Wireless Sensor Networks Amy L. Murphy ITC-IRST, Trento, Italy joint work with Alessandro Giusti, Politecnico.
Network Aware Resource Allocation in Distributed Clouds.
March 6th, 2008Andrew Ofstad ECE 256, Spring 2008 TAG: a Tiny Aggregation Service for Ad-Hoc Sensor Networks Samuel Madden, Michael J. Franklin, Joseph.
1 Pradeep Kumar Gunda (Thanks to Jigar Doshi and Shivnath Babu for some slides) TAG: a Tiny Aggregation Service for Ad-Hoc Sensor Networks Samuel Madden,
TAG: a Tiny Aggregation Service for Ad-Hoc Sensor Networks Authors: Samuel Madden, Michael Franklin, Joseph Hellerstein Presented by: Vikas Motwani CSE.
Benjamin AraiUniversity of California, Riverside Reliable Hierarchical Data Storage in Sensor Networks Song Lin – Benjamin.
ENERGY-EFFICIENT FORWARDING STRATEGIES FOR GEOGRAPHIC ROUTING in LOSSY WIRELESS SENSOR NETWORKS Presented by Prasad D. Karnik.
Lan F.Akyildiz,Weilian Su, Erdal Cayirci,and Yogesh sankarasubramaniam IEEE Communications Magazine 2002 Speaker:earl A Survey on Sensor Networks.
Rushing Attacks and Defense in Wireless Ad Hoc Network Routing Protocols ► Acts as denial of service by disrupting the flow of data between a source and.
Energy-Efficient Monitoring of Extreme Values in Sensor Networks Loo, Kin Kong 10 May, 2007.
Energy-Efficient Shortest Path Self-Stabilizing Multicast Protocol for Mobile Ad Hoc Networks Ganesh Sridharan
Neighborhood-Based Topology Recognition in Sensor Networks S.P. Fekete, A. Kröller, D. Pfisterer, S. Fischer, and C. Buschmann Corby Ziesman.
DATA AGGREGATION Siddhartha Sarkar Roll no: CSE-4 th Year-7 th semester Sensor Networks (CS 704D) Assignment.
1 Shape Segmentation and Applications in Sensor Networks Xianjin Xhu, Rik Sarkar, Jie Gao Department of CS, Stony Brook University INFOCOM 2007.
CountTorrent: Ubiquitous Access to Query Aggregates in Dynamic and Mobile Sensor Networks Abhinav Kamra, Vishal Misra and Dan Rubenstein - Columbia University.
SRL: A Bidirectional Abstraction for Unidirectional Ad Hoc Networks. Venugopalan Ramasubramanian Ranveer Chandra Daniel Mosse.
 Tree in Sensor Network Patrick Y.H. Cheung, and Nicholas F. Maxemchuk, Fellow, IEEE 3 rd New York Metro Area Networking Workshop (NYMAN 2003)
Tufts Wireless Laboratory School Of Engineering Tufts University Paper Review “An Energy Efficient Multipath Routing Protocol for Wireless Sensor Networks”,
By: Gang Zhou Computer Science Department University of Virginia 1 Medians and Beyond: New Aggregation Techniques for Sensor Networks CS851 Seminar Presentation.
1 Constraint-Chaining: On Energy -Efficient Continuous Monitoring in Sensor Networks Adam Silberstein Rebecca Braynard Jun Yang Duke University.
Localized Low-Power Topology Control Algorithms in IEEE based Sensor Networks Jian Ma *, Min Gao *, Qian Zhang +, L. M. Ni *, and Wenwu Zhu +
Load Balanced Link Reversal Routing in Mobile Wireless Ad Hoc Networks Nabhendra Bisnik, Alhussein Abouzeid ECSE Department RPI Costas Busch CSCI Department.
TreeCast: A Stateless Addressing and Routing Architecture for Sensor Networks Santashil PalChaudhuri, Shu Du, Ami K. Saha, and David B. Johnson Department.
1 Low Latency Multimedia Broadcast in Multi-Rate Wireless Meshes Chun Tung Chou, Archan Misra Proc. 1st IEEE Workshop on Wireless Mesh Networks (WIMESH),
Structure-Free Data Aggregation in Sensor Networks.
Directed Diffusion: A Scalable and Robust Communication Paradigm for Sensor Networks Presented by Barath Raghavan.
Introduction Wireless Ad-Hoc Network  Set of transceivers communicating by radio.
TAG: a Tiny AGgregation service for ad-hoc sensor networks Authors: Samuel Madden, Michael J. Franklin, Joseph M. Hellerstein, Wei Hong Presenter: Mingwei.
William Stallings Data and Computer Communications
Introduction to Wireless Sensor Networks
Distributed database approach,
Data-Centric Networking
Presentation transcript:

When processing is cheaper than transmitting Daniel V Uhlig Maryam Rahmaniheris 1

 How to gather interesting data from thousands of Motes? Tens to thousands of motes Unreliable individually  To collect and analyze data Long term low energy deployment Can using processing power at each Mote  Analyze local before sharing data 2

 Transmission of data is expensive compare to CPU cycles 1Kb transmitted 100 meters = 3 million CPU instructions AA power Mote can transmit 1 message per day for about two months (assuming no other power draws) Power density is growing very slowly compared to computation power, storage, etc  Analyze and process locally, only transmitting what is required 3

 Minimize communications ◦ Minimize broadcast/receive time ◦ Minimize message size ◦ Move computations to individual nodes  Nodes pass data in multi-hop fashion towards a root  Select connectivity so graph helps with processing  Handle faulty nodes within network 4

 Max is very simple  What about Count? ◦ Need to avoid double counting due to redundant paths  What about spatial events? ◦ Need to evaluate readings across multiple sensors  Correlation between events  Failures of nodes can loose branches of the tree 6

Connectivity Graph – unstructured or how to structure Diffusion of requests and how to combine data Maintenance messages vs Query messages Reliability of results Load balancing – messages traffic – storage Storage costs at different nodes 7

S.Madden, M.Franklin, J.Hellerstein, and W.Hong Intel Research,

Aggregates values in low power, distributed network Implemented on TinyOS Motes SQL like language to search for values or sets of values – Simple declarative language Energy savings Tree based methodology – Root node generates requests and dissipates down the children 9

Three functions to aggregate results – f (merge function) Each node runs f to combine values =f (, ) EX: =f (, ) – i (initialize function) Generates state record at lowest level of tree EX: – e (evaluator function) Root uses e to generate the final result RESULT=e, EX: SUM/COUNT Functions must be preloaded on Motes or distributed via software protocols 10

Count = Max via tree 11

All searches have different properties that affect aggregate performance Duplicate insensitive – unaffected by double counting (Max, Min) vs (Count, Average) – Restrict network properties Exemplary – return one value (Max/Min) – Sensitive to failure Summary – computation over values (Average) – Less sensitive to failure 12

Distributive – Partial states are the same as final state (Max) Algebraic – Partial states are of fixed size but differ from final state (Average - Sum, Count) Holistic – Partial states contain all sub-records (median) – Unique – similar to Holistic, but partial records may be smaller then holistic Content Sensitive – Size of partial records depend on content (Count Distinct) 13

 Diffusion of requests and then collection of information  Epochs subdivided for each level to complete task ◦ Saves energy ◦ Limits rate of data flow 14

 Snooping – Broadcast messages so others can hear messages ◦ Rejoin tree if parents have failure ◦ Listen to other broadcasts and only broadcast if its values are needed  In case of MAX, do not broadcast if peer has transmitted a higher value  Hypothesis testing – root guesses at value to minimize traffic 15

 Theoretic results for ◦ 2500 Nodes  Savings depend on function  Duplicate Insensitive, summary best ◦ Distributive helps  Holistic is the worse 16

16 Mote network Count number of motes in 4 sec epochs No optimizations Quality of count is due to less radio contention in TAG Centralized used 4685 messages vs TAG’s % reduction, but less then theoretical results – Different loss model, node placement 17

Loss of nodes and subtrees – Maintenance for structured connectivity Single message per node per epoch – Message size might increase at higher level nodes – Root gets overload (Does it always matter?) Epochs give a method for idling nodes – Snooping not included, timing issues 18

 Continuous aggregation ◦ Nodes constantly passing data towards aggregation points  Root free ◦ Any node start query  Query can take different paths ◦ Balances load between nodes  What costs, advantages over TAG? 19

S.Nath, P.Gibbons, S.Seshan, Z.Anderson Microsoft Research,

 TAG ◦ Not robust against node or link failure ◦ A single node failure leads to loss of the entire sub branch's data  Synopsis Diffusion ◦ Exploiting the broadcast nature of wireless medium to enhance reliability ◦ Separating routing from aggregation ◦ The final aggregated data at the sink is independent of the underlying routing topology ◦ Synopsis diffusion can be used on top of any routing structure ◦ The order of evaluations and the number of times each data included in the result is irrelevant 21

Not robust against node or link failure Count = 10

 Multi-path routing ◦ Benefits  Robust  Energy-efficient ◦ Challenges  Duplicate sensitivity  Order sensitivity Count =

 A novel aggregation framework ◦ ODI synopsis: small-sized digest of the partial results  Bit-vectors  Sample  Histogram  Better aggregation topologies ◦ Multi-path routing ◦ Implicit acknowledgment ◦ Adaptive rings  Example aggregates  Performance evaluation 24

 The exact definition of these functions depend on the particular aggregation function: ◦ SG(.)  Takes a sensor reading and generates a synopsis ◦ SF(.,.)  Takes two synopsis and generates a new one ◦ SE(.)  Translates a synopsis into the final answer 25 SG: Synopsis Generation SF: Synopsis Fusion SE: Synopsis Evaluation

 Distribution phase ◦ The aggregate query is flooded ◦ The aggregate topology is constructed  Aggregation phase ◦ Aggregated values are routed toward Sink ◦ SG() and SF() functions are used to create partial results 26

 The sink is in R0  A node is in Ri if it’s i hops away from sink  Nodes in Ri-1 should hear the broadcast by nodes in Ri  Loose synchronization between nodes in different rings  Each node transmits only once ◦ Energy cost same as tree 27 R3R3 R2R2 R0R0 R1R1 A B C

 Coin tossing experiment CT(x) used in Flajolet and Martin’s Algorithm: ◦ For i=1,…,x-1: CT(x) = i with probability ◦ Simulates the behavior of the exponential hash function ◦ Synopsis: a bit vector of length k > log(n)  n is an upper bound on the number of the sensor nodes in the network ◦ SG(): a bit vector of length k with only the CT(k)th bit is set ◦ SF(): bit wise Boolean OR ◦ SE(): the index of lowest-order 0 in the bit vector= i-> 28 SG: Synopsis Generation SF: Synopsis Fusion SE: Synopsis Evaluation Magic Constant

 The number of live sensor nodes, N, is proportional to Count 1 bits 4 29 Intuition : The probability of N nodes all failing to set the i th bit is which is approximately 0.37 when and even smaller for larger N. SG: Synopsis Generation SF: Synopsis Fusion SE: Synopsis Evaluation

Aggregation DAGCanonical left-deep tree SG SF r1r2r5r3r4 s SG r1r2 r3 r4 r5 SF s 30 SG: Synopsis Generation SF: Synopsis Fusion SE: Synopsis Evaluation

◦ P1: SG() preserves duplicates  If two reading are considered duplicates then the same synopsis is generated ◦ P2: SF() is commutative  SF(s1, s2) = SF(s2, s1) ◦ P3: SF() is associative  SF(s1, SF(s2, s3)) = SF(SF(s1, s2), s3) ◦ P4: SF() is same-synopsis idempotent  SF(s, s) = s Theorem: Properties P1-P4 are necessary and sufficient properties for ODI-Correctness 31

 Uniform Sample of Readings ◦ Synopsis: A sample of size K of tuples ◦ SG(): Output the tuple ◦ SF(s,s’): outputs the K tuples in s∪s’ with the K largest r i ◦ SE(s): Output the set of values val i in s ◦ Useful holistic aggregation 32 SG: Synopsis Generation SF: Synopsis Fusion SE: Synopsis Evaluation

 Frequent Items (items occurring at least T times) ◦ Synopsis: A set of pairs, the values are unique and the weights are at least log(T) ◦ SG(): Compute CT(k) where k>log(n) and call this weight and if it’s at least log(T) output ◦ SF(s,s’): For each distinct value discard all but the pair with maximum weight. Output the remaining pairs. ◦ SE(s): Output for each pair in s as a frequent value and its approximate count ◦ Intuition: A value occurring at least T time is expected to have at least one of its calls to CT() return at least log(T)  p=1/T 33 SG: Synopsis Generation SF: Synopsis Fusion SE: Synopsis Evaluation

 Communication error ◦ 1-Percent contributing ◦ h: height of DAG ◦ k: the number of neighbors each nodes has ◦ p: probability of loss ◦ The overall communication error upper bound: ◦ If p=0.1, h=10 then the error is negligible with k=3  Approximation error ◦ Introduced by SG(), SF(), and SE() functions ◦ Theorem 2: any approximation error guarantees provided for the centralized data stream scenario immediately applies to a synopsis diffusion algorithm, as long as the data stream synopsis is ODI-correct. 34

 Implicit acknowledgement provided by ODI synopses ◦ Retransmission  High energy cost and delay ◦ Adapting the topology  When the number of times a node’s transmission is included in the parents transmission is below a threshold  Assigning the node to a ring that can have a good number of parents  Assign a node in ring i with probability p to :  Ring i +1 If  ni > ni-1  ni+1 > ni -1 and ni+2 > ni  Ring i -1 If  ni-2 > ni-1  ni-1 ni 35

RingsAdaptive Rings 36

 The algorithms are implemented in TAG simulator  600 sensors deployed randomly in a 20 ft * 20 ft grid  The query node is in the center  Loss probabilities are assigned based of the distance between nodes 37

RMS Error% Value Included 38

 Pros ◦ High reliability and robustness ◦ More accurate answers ◦ Implicit acknowledgment ◦ Dynamic topology adaptation ◦ Moderately affected by mobility  Cons ◦ Approximation error ◦ Low node density decreases the benefits ◦ The fusion functions should be defined for each aggregation function ◦ Increased message size 39

 Is there any benefit in coupling routing with aggregation? ◦ Choosing the paths and finding the optimal aggregation points ◦ Routing the sensed data along a longer path to maximize aggregation ◦ Finding the optimal routing structure  Considering energy cost of links  NP-Complete  Heuristics (Greedy Incremental)  Considering data correlation in the aggregation process ◦ Spatial ◦ Temporal  Defining a threshold  TiNA 40

 Could energy saving gained by aggregation be outweighed by the cost of it? ◦ Aggregation function cost  Storage cost  Computation cost (Number of CPU cycles)  No mobility ◦ Static aggregation tree  Structure-less or structured? That is the question… ◦ Continuous ◦ On-demand 41

 Transmitting large amounts of data on the internet is slow ◦ Better to process locally and transmit the interesting parts only 42

 How does query rate affect design decisions?  Load balancing between levels of the tree ◦ Overload root and main nodes  How will video capabilities of Imote affect aggregation models? 43

44

 Query can originate at any node, not just the root  Histogram data so different levels of the tree hold different details of data. ◦ Child hold wider range/smaller area ◦ Parents hold smaller range / wider area 45

 Avoid bottlenecks ◦ Queue can originate anywhere  Avoid overload the one root node ◦ Different nodes can answer different questions quickly  Must constantly aggregating data 46