SlideShare ist ein Scribd-Unternehmen logo
1 von 40
Scalable membership management
                 and
         failure detection?

           Vinay Setty
            INF5360
What is Gossiping?
• Spread of information in a random manner
• Some examples:
  – Human gossiping
  – Epidemic diseases
  – Physical phenomenon: wild fire, diffusion etc
  – Computer viruses and worms
Gossiping in Computer Science
• Term first coined by Demers et al (1987)
• Some applications of gossip protocols
  – Peer Sampling
  – Data Aggregation
  – Clustering
  – Information Dissemination (Multicast, Pub/Sub)
  – Overlay/topology
  – Maintenance
  – Failure detection?
Gossip-Based Protocol: Example


             3       4
     0

                         5


         1
                 2
 9
                                 8

                             7
             6
Today’s Focus
• Theoretical angle for Gossip-based protocols
  [Allavena et al PODC 2005]
  – Probability of partitioning
  – Time till partitioning
  – Bounds on in-degree
  – Essential elements of gossiping
  – Simulation results
• Cyclon [Voulgaris et al]
• Scamp [Ganesh et al]
• NewsCast [Jelasity et al]
Membership Service
• Full Membership
  – Complete knowledge at each node
  – Random subset used for gossiping
  – Not scalable
  – Hard to maintain
• Partial Membership
  – Random subset at each node
  – Gossip partners chosen from local view
View Selection
                 L1
        s         s,p,r   p

            v s,p,t u              r

    t             t,q,r
                              q
                  L2

                                  L1                     L2
v




                                       Weighted with w
Essential Elements of Gossiping
• Mixing: Construct a list L1 consisting of local
  views of local view of node u
  – Guarantees non partitioning
  – “Pull” based
• Reinforcement: Construct a list L2 consisting
  of local views of nodes that requested local
  view of u
  – Balances network
  – removes old possibly dead edges, adds new edges
Partitioning and Size Estimate




• A and B partition iff x=1 and y=0
• Partitioning is least possible when x=y
• Goal of protocol is to maintain this balance
Size Estimates
• Idea:
  – Assuming edges were drawn uniformly randomly,
    expected x+y  |A|
  – x is estimate of size of A by nodes in A
  – y is estimate of size of A by nodes in B
• Mixing:
  – Agreeing on estimation of x and y ensures no
    partition (even if x and y are not accurate)
• Reinforcement:
  – Brings estimation of x and y to correct value
K-regularity
•   View Size: k
•   Number of nodes: n
•   Fraction of nodes in partition: γ
•   |A|= γn ≤ |B|
•   #edges from A to B: (1-x)γkn
•   #edges from B to A: y (1-γ)kn
•   Number of edges in A-B cut:
    – (1-x)γkn +x (1-γ)kn              (since x=y)
    – ≥ γkn                         (assuming γ≤½)
Time Till Partitioning
•   View Size: k
•   Number of nodes: n
•   Fraction of nodes in partition: γ
•   Churn rate: μ (μn nodes leave and join)
•   Claim: Expected time before a partition of size
    γ happens ≈ 2γkn
    – As long as μ≪γkn
100, 000 nodes, view sizes of 17, a fanout of 3, and a loosely
                                   synchronised syst em, t he maximum in-degree was always
    which re-samples ran-          below 4.5 t imes t hat of a random graph and t he st andard
g t he names of t he nodes         deviat ion was not more t han 3.2 t imes larger t han t hat of a
                    Iterations until Partitioning
 y or anot her is doomed
enat ion of all t he views.
                                   random graph. T hese values would improve wit h increased
                                   fanout , but even a fanout of 2 gives sat isfact ory perfor-
 onds t o creat ing a new          mance.
eplacement from t he old
om V at each it erat ion.
                                                           10000
annot reappear wit hout
 cement . T he diversit y                                   9000
  ime, and in fact rat her                                  8000
                                    Number of iterations
 rk. Not e t hat it is t he-                                7000
e by creat ing a protnodes: n
          Number of ocol
                                                            6000
 t at ion on V, size: tk = log n
          View but his is
          Churn: n/32
esn’t necessarily behave                                    5000
ving or joining t he net -                                  4000
 ively add t he names of                                    3000
  t o V, a process we call
ome reinforcement , even                                    2000
 n t he art icle: each pro-                                 1000
  hen sending a message.                                       0
 nd Cyclon [16] as well:                                           1   1.5       2       2.5      3     3.5
eir view t hat t hey t hen                                               Log10 of the number of nodes

aviour in say t he cont ext           F igure 4: N umber of it erat ions unt il part it ioning
 t he “ News Event s” are
des. Let only t he nodes             We were int erest ed in mat ching our t heoret ical result s
odes add t heir names t o          about part it ioning and churn. We ran simulat ions evaluat -
Event s” inst ead of every         ing t he number of it erat ions unt il part it ioning. By part it ion-
View Size vs Time until Partition


Number of nodes: n
View size: k = log n
Churn: n/32
Simplified Model for Proof
– Single randomly chosen element from view is
  replaced instead of whole views
– Assumption: The out-edges of nodes of A are
  identically distributed and same applies to B
– a = #edges from A to A
– c = #edges from A to B
– b = #edges from B to A
– d = #edges from B to B
Proof Intuition
Partition state: a = γkn and b = 0
In-Degree Analysis
• Load balancing requires balance in in-degree
  distribution
• In-degree is governed by the way edges created,
  copied and destroyed
• Copying some edges more than others cause
  variability in in-degree
• Node living longer is expected to have higher in-
  degree
• Solution: Increase reinforcement and keep track
  of timestamps like in Cyclon
• Simulation: max in-degree < 4.5 times of random
  graph and standard deviation < 3.2 times
Discussion
• Are these theoretical guarantees practically
  useful?
• Goal is not provide failure detection
Cyclon
• Consists of same elements as suggested by
  [Allavena et al PODC 2005]
• [Allavena et al PODC 2005] Analysis holds for
  Cyclon
• Major differences:
  – Timestamps
  – shuffling
Basic Shuffling
• Select a random subset of l neighbors (1 ≤ l ≤ c) from P’s
  own cache, and a random peer, Q, within this subset,
  where l is a system parameter, called shuffle length.
• Replace Q’s address with P’s address.
• Send the updated subset to Q.
• Receive from Q a subset of no more than l of Q’s neighbors.
• Discard entries pointing to P, and entries that are already in
  P’s cache.
• Update P’s cache to include all remaining entries, by
   – firstly using empty cache slots (if any), and
   – secondly replacing entries among the ones originally sent to Q.
Shuffling Example
Enhanced Shuffling
• Increase by one the age of all neighbors.
• Select neighbor Q with the highest age among all neighbors, and l −
  1 other random neighbors.
• Replace Q’s entry with a new entry of age 0 and with P’s address.
• Send the updated subset to peer Q.
• Receive from Q a subset of no more that l of its own entries.
• Discard entries pointing at P and entries already contained in P’s
  cache.
• Update P’s cache to include all remaining entries, by firstly using
  empty
• cache slots (if any), and secondly replacing entries among the ones
  sent to Q.
Time Until Dead Links Removed
removed. Note that the number of clusters decreases as we approach 100% node
removal because the total number of surviving nodes becomes too small. Fig-

                    Number of Clusters
ure 7(b) shows the number of nodes not belonging to the largest cluster, in log
scale.
      These graphs show considerable robustness to node failures, especially con-
sidering the fact that in the early stages of clustering very few nodes are out of
the largest cluster, which indicates that most nodes are still connected in a single




Fig. 7. (a) Number of disjoint clusters, as a result of removing a large percentage of nodes. Shows
that the overlay does not break into two or more disjoint clusters, unless a major percentage of the
nodes are removed. (b) Number of nodes not belonging to the largest cluster. Shows that in the first
steps of clustering only a few nodes are separated from the main cluster, which still connects the
Tolerance to Partitioning
In-Degree Distribution
SCAMP
• Partial knowledge of the membership: local
  view
• Fanout automatically set = size of the local
  view
• Fanout evolves naturally with the size of the
  group
  – Size of local views converges towards C.log(n)
Join (Subscription)

Subscription to          Subscription forwarded
a random member                                             P=1/sizeof view
                               s                  1           s
S                    0                                               (1-P)
                                    s
                                                      s   P=1/sizeof view

                                     s            2            s
                                                                     (1-P)

                                                      s
                                                          P=1/sizeof view

                                                  3           s
                                                                    (1-P)
Join(Subscription) algorithm

                                 7 6
Local view       1 4 5   6       4
                     0
                             6
                                 7 2              3 6
                             6            6
                                 1                2
                 6

             0
                                  8 3 6           7 0 1       5 6
                                              6           6
                                 5                 8          7
Load Balancing
• Indirection:
  – Forward the subscription instead of handling
    request
• Lease associated with each subscription
• Periodically nodes have to re-subscribe
  – Nodes having failed permanently will time out
  – Re-balance the partial views
Unsubscription

Local view                                8 9 0   8 9 4
   1 4 5     Unsub (0), [1,4,5]
                                      4     x       x
     0

                                          7 3 0   7 3 5
                                      1     y       y


                                          6 0 2   6 0 1
                                  5
                                            z       z
Degree

• System modelled as random directed graph
• D(N) = Average out-degree for N-nodes
  system
• Subscription adds D(N)+1 directed arcs, so
• (N+1) D(N+1) = N D(N) + D(N)+1
• Solution of this recursion is
• D(N)=D(1)+1/2+1/3+…+1/N  Log(N)
Distribution of view size

                  35000

                  30000
                                                                    Log=13.12
                  25000
Number of nodes




                  20000

                  15000

                  10000                                             Log=12.2

                   5000

                      0
                          0   5    10    15    20       25   30     35    40    45    50
                                                    View Size

                                  200 000 Node System           500 000 Node System
                                                                                           33
Reliability: 5000 node system
                  1

               0.98

               0.96
 Reliability




               0.94

               0.92

                0.9
                      0   500       1000      1500      2000        2500
                                 Number of failures


                            SCAMP
                            Global membership knowledge, fanout=8
                            Global membership knowledge, fanout=9

                                                                           34
NewsCast
• Goal: Aggregate information in
  – a large and dynamic
  – distributed environment
  – a robust and dependable manner
Idea
• Gets news from application, timestamps it and
  adds local peer address to the cache entry
• Finds a random peer in cache addresses
  – Sends all cache entries to this peer
  – Receives all cache entries from that peer
• Passes on cache entries (containing news items)
  to application
• Merges old cache with received cache
  – Keeps at most C most recent cache entries
Aggregation
• Each node ni maintains a single number xi
• Every node ni selects a random node nk, and
  sends its value xi to nk
• nk responds with the aggregate (e.g. max(xi,xk)
  ) of the incoming and its own value
• 4. Aggregate values will converge
  “exponentially”
Path length under failures
Connectivity Under Failures
Aggregation
                                      1
proportion of not-reached nodes



                                     0.1



                                    0.01



                                   0.001
                                                  theoretical model
                                                  c=20
                                                  c=40
                                                  c=80
                                  0.0001
                                           6      7          8           9   10   11
                                                                 cycle

Weitere ähnliche Inhalte

Was ist angesagt?

Dynamic Compression Transmission for Energy Harvesting Networks
Dynamic Compression Transmission for Energy Harvesting NetworksDynamic Compression Transmission for Energy Harvesting Networks
Dynamic Compression Transmission for Energy Harvesting NetworksCristiano Tapparello
 
Meta online learning: experiments on a unit commitment problem (ESANN2014)
Meta online learning: experiments on a unit commitment problem (ESANN2014)Meta online learning: experiments on a unit commitment problem (ESANN2014)
Meta online learning: experiments on a unit commitment problem (ESANN2014)Jialin LIU
 
STRIP: stream learning of influence probabilities.
STRIP: stream learning of influence probabilities.STRIP: stream learning of influence probabilities.
STRIP: stream learning of influence probabilities.Albert Bifet
 
Sampling with Halton Points on n-Sphere
Sampling with Halton Points on n-SphereSampling with Halton Points on n-Sphere
Sampling with Halton Points on n-SphereDanny Luk
 
EuroPython 2017 - PyData - Deep Learning your Broadband Network @ HOME
EuroPython 2017 - PyData - Deep Learning your Broadband Network @ HOMEEuroPython 2017 - PyData - Deep Learning your Broadband Network @ HOME
EuroPython 2017 - PyData - Deep Learning your Broadband Network @ HOMEHONGJOO LEE
 
Yampa AFRP Introduction
Yampa AFRP IntroductionYampa AFRP Introduction
Yampa AFRP IntroductionChengHui Weng
 
Alex Smola, Professor in the Machine Learning Department, Carnegie Mellon Uni...
Alex Smola, Professor in the Machine Learning Department, Carnegie Mellon Uni...Alex Smola, Professor in the Machine Learning Department, Carnegie Mellon Uni...
Alex Smola, Professor in the Machine Learning Department, Carnegie Mellon Uni...MLconf
 
Aaex7 group2(中英夾雜)
Aaex7 group2(中英夾雜)Aaex7 group2(中英夾雜)
Aaex7 group2(中英夾雜)Shiang-Yun Yang
 
Speaker Diarization
Speaker DiarizationSpeaker Diarization
Speaker DiarizationHONGJOO LEE
 
Practical Spherical Harmonics Based PRT Methods
Practical Spherical Harmonics Based PRT MethodsPractical Spherical Harmonics Based PRT Methods
Practical Spherical Harmonics Based PRT MethodsNaughty Dog
 
Wireless Localization: Ranging (first part)
Wireless Localization: Ranging (first part)Wireless Localization: Ranging (first part)
Wireless Localization: Ranging (first part)Stefano Severi
 
End of Sprint 5
End of Sprint 5End of Sprint 5
End of Sprint 5dm_work
 
Advanced Topics Sorting
Advanced Topics SortingAdvanced Topics Sorting
Advanced Topics SortingSri Prasanna
 
Discrete Logarithmic Problem- Basis of Elliptic Curve Cryptosystems
Discrete Logarithmic Problem- Basis of Elliptic Curve CryptosystemsDiscrete Logarithmic Problem- Basis of Elliptic Curve Cryptosystems
Discrete Logarithmic Problem- Basis of Elliptic Curve CryptosystemsNIT Sikkim
 

Was ist angesagt? (20)

Dynamic Compression Transmission for Energy Harvesting Networks
Dynamic Compression Transmission for Energy Harvesting NetworksDynamic Compression Transmission for Energy Harvesting Networks
Dynamic Compression Transmission for Energy Harvesting Networks
 
Realtime Analytics
Realtime AnalyticsRealtime Analytics
Realtime Analytics
 
Meta online learning: experiments on a unit commitment problem (ESANN2014)
Meta online learning: experiments on a unit commitment problem (ESANN2014)Meta online learning: experiments on a unit commitment problem (ESANN2014)
Meta online learning: experiments on a unit commitment problem (ESANN2014)
 
STRIP: stream learning of influence probabilities.
STRIP: stream learning of influence probabilities.STRIP: stream learning of influence probabilities.
STRIP: stream learning of influence probabilities.
 
Sampling with Halton Points on n-Sphere
Sampling with Halton Points on n-SphereSampling with Halton Points on n-Sphere
Sampling with Halton Points on n-Sphere
 
EuroPython 2017 - PyData - Deep Learning your Broadband Network @ HOME
EuroPython 2017 - PyData - Deep Learning your Broadband Network @ HOMEEuroPython 2017 - PyData - Deep Learning your Broadband Network @ HOME
EuroPython 2017 - PyData - Deep Learning your Broadband Network @ HOME
 
Yampa AFRP Introduction
Yampa AFRP IntroductionYampa AFRP Introduction
Yampa AFRP Introduction
 
Alex Smola, Professor in the Machine Learning Department, Carnegie Mellon Uni...
Alex Smola, Professor in the Machine Learning Department, Carnegie Mellon Uni...Alex Smola, Professor in the Machine Learning Department, Carnegie Mellon Uni...
Alex Smola, Professor in the Machine Learning Department, Carnegie Mellon Uni...
 
Aaex7 group2(中英夾雜)
Aaex7 group2(中英夾雜)Aaex7 group2(中英夾雜)
Aaex7 group2(中英夾雜)
 
Matched filter
Matched filterMatched filter
Matched filter
 
Speaker Diarization
Speaker DiarizationSpeaker Diarization
Speaker Diarization
 
Europy17_dibernardo
Europy17_dibernardoEuropy17_dibernardo
Europy17_dibernardo
 
ECE 565 FInal Project
ECE 565 FInal ProjectECE 565 FInal Project
ECE 565 FInal Project
 
Practical Spherical Harmonics Based PRT Methods
Practical Spherical Harmonics Based PRT MethodsPractical Spherical Harmonics Based PRT Methods
Practical Spherical Harmonics Based PRT Methods
 
Recurrent Instance Segmentation (UPC Reading Group)
Recurrent Instance Segmentation (UPC Reading Group)Recurrent Instance Segmentation (UPC Reading Group)
Recurrent Instance Segmentation (UPC Reading Group)
 
Dsp manual
Dsp manualDsp manual
Dsp manual
 
Wireless Localization: Ranging (first part)
Wireless Localization: Ranging (first part)Wireless Localization: Ranging (first part)
Wireless Localization: Ranging (first part)
 
End of Sprint 5
End of Sprint 5End of Sprint 5
End of Sprint 5
 
Advanced Topics Sorting
Advanced Topics SortingAdvanced Topics Sorting
Advanced Topics Sorting
 
Discrete Logarithmic Problem- Basis of Elliptic Curve Cryptosystems
Discrete Logarithmic Problem- Basis of Elliptic Curve CryptosystemsDiscrete Logarithmic Problem- Basis of Elliptic Curve Cryptosystems
Discrete Logarithmic Problem- Basis of Elliptic Curve Cryptosystems
 

Andere mochten auch

Seara YP - Marketing Turistic - TodayYouCanDoAnything
Seara YP - Marketing Turistic - TodayYouCanDoAnythingSeara YP - Marketing Turistic - TodayYouCanDoAnything
Seara YP - Marketing Turistic - TodayYouCanDoAnythingIoana Sigarteu
 
A2 Advanced Portfolio Evaluation (question one)
A2 Advanced Portfolio Evaluation (question one)A2 Advanced Portfolio Evaluation (question one)
A2 Advanced Portfolio Evaluation (question one)hevaw
 
Seara YP - Marketing Turistic - Travelers of Bucharest
Seara YP - Marketing Turistic - Travelers of BucharestSeara YP - Marketing Turistic - Travelers of Bucharest
Seara YP - Marketing Turistic - Travelers of BucharestIoana Sigarteu
 
nSIGHT Presentation
nSIGHT PresentationnSIGHT Presentation
nSIGHT PresentationOHRNIH
 
Evaluation question 2
Evaluation question 2Evaluation question 2
Evaluation question 2hevaw
 
Prezentare Oana Turturica - Hai cu Bicla - Seara YP Marketingul pentru sportu...
Prezentare Oana Turturica - Hai cu Bicla - Seara YP Marketingul pentru sportu...Prezentare Oana Turturica - Hai cu Bicla - Seara YP Marketingul pentru sportu...
Prezentare Oana Turturica - Hai cu Bicla - Seara YP Marketingul pentru sportu...Ioana Sigarteu
 
A practical guide to analytical method validation, including measurement unce...
A practical guide to analytical method validation, including measurement unce...A practical guide to analytical method validation, including measurement unce...
A practical guide to analytical method validation, including measurement unce...Victor Huamaní León
 
Chapter 15 top 10 - designing and managing integrated marketing concepts
Chapter 15 top 10 - designing and managing integrated marketing conceptsChapter 15 top 10 - designing and managing integrated marketing concepts
Chapter 15 top 10 - designing and managing integrated marketing conceptsk3llycr1s0st
 
Chap 5 - creating customer value, satisfaction and loyalty
Chap 5 - creating customer value, satisfaction and loyaltyChap 5 - creating customer value, satisfaction and loyalty
Chap 5 - creating customer value, satisfaction and loyaltyk3llycr1s0st
 
Coldplay
ColdplayColdplay
Coldplayhevaw
 
Business process outsourcing - Entrep Report AGSB 2012
Business process outsourcing - Entrep Report AGSB 2012Business process outsourcing - Entrep Report AGSB 2012
Business process outsourcing - Entrep Report AGSB 2012k3llycr1s0st
 
Seara YP - Marketing Turistic - ImperatorTravel.ro
Seara YP - Marketing Turistic - ImperatorTravel.roSeara YP - Marketing Turistic - ImperatorTravel.ro
Seara YP - Marketing Turistic - ImperatorTravel.roIoana Sigarteu
 
A fine frenzy
A fine frenzyA fine frenzy
A fine frenzyhevaw
 
Chap 10 – Crafting the Brand Positioning
Chap 10 – Crafting the Brand PositioningChap 10 – Crafting the Brand Positioning
Chap 10 – Crafting the Brand Positioningk3llycr1s0st
 
Chapter 5 creating customer value, satisfaction and loyalty
Chapter 5  creating customer value, satisfaction and loyaltyChapter 5  creating customer value, satisfaction and loyalty
Chapter 5 creating customer value, satisfaction and loyaltyk3llycr1s0st
 

Andere mochten auch (16)

Seara YP - Marketing Turistic - TodayYouCanDoAnything
Seara YP - Marketing Turistic - TodayYouCanDoAnythingSeara YP - Marketing Turistic - TodayYouCanDoAnything
Seara YP - Marketing Turistic - TodayYouCanDoAnything
 
A2 Advanced Portfolio Evaluation (question one)
A2 Advanced Portfolio Evaluation (question one)A2 Advanced Portfolio Evaluation (question one)
A2 Advanced Portfolio Evaluation (question one)
 
Seara YP - Marketing Turistic - Travelers of Bucharest
Seara YP - Marketing Turistic - Travelers of BucharestSeara YP - Marketing Turistic - Travelers of Bucharest
Seara YP - Marketing Turistic - Travelers of Bucharest
 
nSIGHT Presentation
nSIGHT PresentationnSIGHT Presentation
nSIGHT Presentation
 
Evaluation question 2
Evaluation question 2Evaluation question 2
Evaluation question 2
 
Prezentare Oana Turturica - Hai cu Bicla - Seara YP Marketingul pentru sportu...
Prezentare Oana Turturica - Hai cu Bicla - Seara YP Marketingul pentru sportu...Prezentare Oana Turturica - Hai cu Bicla - Seara YP Marketingul pentru sportu...
Prezentare Oana Turturica - Hai cu Bicla - Seara YP Marketingul pentru sportu...
 
Campioni speciali
Campioni specialiCampioni speciali
Campioni speciali
 
A practical guide to analytical method validation, including measurement unce...
A practical guide to analytical method validation, including measurement unce...A practical guide to analytical method validation, including measurement unce...
A practical guide to analytical method validation, including measurement unce...
 
Chapter 15 top 10 - designing and managing integrated marketing concepts
Chapter 15 top 10 - designing and managing integrated marketing conceptsChapter 15 top 10 - designing and managing integrated marketing concepts
Chapter 15 top 10 - designing and managing integrated marketing concepts
 
Chap 5 - creating customer value, satisfaction and loyalty
Chap 5 - creating customer value, satisfaction and loyaltyChap 5 - creating customer value, satisfaction and loyalty
Chap 5 - creating customer value, satisfaction and loyalty
 
Coldplay
ColdplayColdplay
Coldplay
 
Business process outsourcing - Entrep Report AGSB 2012
Business process outsourcing - Entrep Report AGSB 2012Business process outsourcing - Entrep Report AGSB 2012
Business process outsourcing - Entrep Report AGSB 2012
 
Seara YP - Marketing Turistic - ImperatorTravel.ro
Seara YP - Marketing Turistic - ImperatorTravel.roSeara YP - Marketing Turistic - ImperatorTravel.ro
Seara YP - Marketing Turistic - ImperatorTravel.ro
 
A fine frenzy
A fine frenzyA fine frenzy
A fine frenzy
 
Chap 10 – Crafting the Brand Positioning
Chap 10 – Crafting the Brand PositioningChap 10 – Crafting the Brand Positioning
Chap 10 – Crafting the Brand Positioning
 
Chapter 5 creating customer value, satisfaction and loyalty
Chapter 5  creating customer value, satisfaction and loyaltyChapter 5  creating customer value, satisfaction and loyalty
Chapter 5 creating customer value, satisfaction and loyalty
 

Ähnlich wie Scalable membership management

Thesis : &quot;IBBET : In Band Bandwidth Estimation for LAN&quot;
Thesis : &quot;IBBET : In Band Bandwidth Estimation for LAN&quot;Thesis : &quot;IBBET : In Band Bandwidth Estimation for LAN&quot;
Thesis : &quot;IBBET : In Band Bandwidth Estimation for LAN&quot;Vishalkumarec
 
Clustering of graphs and search of assemblages
Clustering of graphs and search of assemblagesClustering of graphs and search of assemblages
Clustering of graphs and search of assemblagesData-Centric_Alliance
 
convolutional_neural_networks in deep learning
convolutional_neural_networks in deep learningconvolutional_neural_networks in deep learning
convolutional_neural_networks in deep learningssusere5ddd6
 
Massively Parallel K-Nearest Neighbor Computation on Distributed Architectures
Massively Parallel K-Nearest Neighbor Computation on Distributed Architectures Massively Parallel K-Nearest Neighbor Computation on Distributed Architectures
Massively Parallel K-Nearest Neighbor Computation on Distributed Architectures Intel® Software
 
Change Point Analysis
Change Point AnalysisChange Point Analysis
Change Point AnalysisMark Conway
 
Optimizing Data Partitioning at Broadcasting the Data
Optimizing Data Partitioning at Broadcasting the DataOptimizing Data Partitioning at Broadcasting the Data
Optimizing Data Partitioning at Broadcasting the DataTakashi Yamanoue
 
Concurrency in Distributed Systems : Leslie Lamport papers
Concurrency in Distributed Systems : Leslie Lamport papersConcurrency in Distributed Systems : Leslie Lamport papers
Concurrency in Distributed Systems : Leslie Lamport papersSubhajit Sahu
 
Financial Networks III. Centrality and Systemic Importance
Financial Networks III. Centrality and Systemic ImportanceFinancial Networks III. Centrality and Systemic Importance
Financial Networks III. Centrality and Systemic ImportanceKimmo Soramaki
 
A network pruning based approach for subset specific influential detection
A network pruning based approach for subset specific influential detectionA network pruning based approach for subset specific influential detection
A network pruning based approach for subset specific influential detectionArun Kalyanasundaram
 
Lecture 7: Data-Intensive Computing for Text Analysis (Fall 2011)
Lecture 7: Data-Intensive Computing for Text Analysis (Fall 2011)Lecture 7: Data-Intensive Computing for Text Analysis (Fall 2011)
Lecture 7: Data-Intensive Computing for Text Analysis (Fall 2011)Matthew Lease
 
PixelCNN, Wavenet, Normalizing Flows - Santiago Pascual - UPC Barcelona 2018
PixelCNN, Wavenet, Normalizing Flows - Santiago Pascual - UPC Barcelona 2018PixelCNN, Wavenet, Normalizing Flows - Santiago Pascual - UPC Barcelona 2018
PixelCNN, Wavenet, Normalizing Flows - Santiago Pascual - UPC Barcelona 2018Universitat Politècnica de Catalunya
 
Markov Chain Monitoring - Application to demand prediction in bike sharing sy...
Markov Chain Monitoring - Application to demand prediction in bike sharing sy...Markov Chain Monitoring - Application to demand prediction in bike sharing sy...
Markov Chain Monitoring - Application to demand prediction in bike sharing sy...Harshal Chaudhari
 
Linear Size Meshes
Linear Size MeshesLinear Size Meshes
Linear Size MeshesDon Sheehy
 
エンドツーエンド音声合成に向けたNIIにおけるソフトウェア群 ~ TacotronとWaveNetのチュートリアル (Part 1)~
エンドツーエンド音声合成に向けたNIIにおけるソフトウェア群 ~ TacotronとWaveNetのチュートリアル (Part 1)~エンドツーエンド音声合成に向けたNIIにおけるソフトウェア群 ~ TacotronとWaveNetのチュートリアル (Part 1)~
エンドツーエンド音声合成に向けたNIIにおけるソフトウェア群 ~ TacotronとWaveNetのチュートリアル (Part 1)~Yamagishi Laboratory, National Institute of Informatics, Japan
 
The Power of Motif Counting Theory, Algorithms, and Applications for Large Gr...
The Power of Motif Counting Theory, Algorithms, and Applications for Large Gr...The Power of Motif Counting Theory, Algorithms, and Applications for Large Gr...
The Power of Motif Counting Theory, Algorithms, and Applications for Large Gr...Nesreen K. Ahmed
 
Convolutional Neural Networks - Veronica Vilaplana - UPC Barcelona 2018
Convolutional Neural Networks - Veronica Vilaplana - UPC Barcelona 2018Convolutional Neural Networks - Veronica Vilaplana - UPC Barcelona 2018
Convolutional Neural Networks - Veronica Vilaplana - UPC Barcelona 2018Universitat Politècnica de Catalunya
 
Queuing theory and traffic analysis in depth
Queuing theory and traffic analysis in depthQueuing theory and traffic analysis in depth
Queuing theory and traffic analysis in depthIdcIdk1
 
Lecture 02 internet video search
Lecture 02 internet video searchLecture 02 internet video search
Lecture 02 internet video searchzukun
 

Ähnlich wie Scalable membership management (20)

Thesis : &quot;IBBET : In Band Bandwidth Estimation for LAN&quot;
Thesis : &quot;IBBET : In Band Bandwidth Estimation for LAN&quot;Thesis : &quot;IBBET : In Band Bandwidth Estimation for LAN&quot;
Thesis : &quot;IBBET : In Band Bandwidth Estimation for LAN&quot;
 
Clustering of graphs and search of assemblages
Clustering of graphs and search of assemblagesClustering of graphs and search of assemblages
Clustering of graphs and search of assemblages
 
CNN for modeling sentence
CNN for modeling sentenceCNN for modeling sentence
CNN for modeling sentence
 
2019 GDRR: Blockchain Data Analytics - ChainNet: Learning on Blockchain Graph...
2019 GDRR: Blockchain Data Analytics - ChainNet: Learning on Blockchain Graph...2019 GDRR: Blockchain Data Analytics - ChainNet: Learning on Blockchain Graph...
2019 GDRR: Blockchain Data Analytics - ChainNet: Learning on Blockchain Graph...
 
convolutional_neural_networks in deep learning
convolutional_neural_networks in deep learningconvolutional_neural_networks in deep learning
convolutional_neural_networks in deep learning
 
Massively Parallel K-Nearest Neighbor Computation on Distributed Architectures
Massively Parallel K-Nearest Neighbor Computation on Distributed Architectures Massively Parallel K-Nearest Neighbor Computation on Distributed Architectures
Massively Parallel K-Nearest Neighbor Computation on Distributed Architectures
 
Change Point Analysis
Change Point AnalysisChange Point Analysis
Change Point Analysis
 
Optimizing Data Partitioning at Broadcasting the Data
Optimizing Data Partitioning at Broadcasting the DataOptimizing Data Partitioning at Broadcasting the Data
Optimizing Data Partitioning at Broadcasting the Data
 
Concurrency in Distributed Systems : Leslie Lamport papers
Concurrency in Distributed Systems : Leslie Lamport papersConcurrency in Distributed Systems : Leslie Lamport papers
Concurrency in Distributed Systems : Leslie Lamport papers
 
Financial Networks III. Centrality and Systemic Importance
Financial Networks III. Centrality and Systemic ImportanceFinancial Networks III. Centrality and Systemic Importance
Financial Networks III. Centrality and Systemic Importance
 
A network pruning based approach for subset specific influential detection
A network pruning based approach for subset specific influential detectionA network pruning based approach for subset specific influential detection
A network pruning based approach for subset specific influential detection
 
Lecture 7: Data-Intensive Computing for Text Analysis (Fall 2011)
Lecture 7: Data-Intensive Computing for Text Analysis (Fall 2011)Lecture 7: Data-Intensive Computing for Text Analysis (Fall 2011)
Lecture 7: Data-Intensive Computing for Text Analysis (Fall 2011)
 
PixelCNN, Wavenet, Normalizing Flows - Santiago Pascual - UPC Barcelona 2018
PixelCNN, Wavenet, Normalizing Flows - Santiago Pascual - UPC Barcelona 2018PixelCNN, Wavenet, Normalizing Flows - Santiago Pascual - UPC Barcelona 2018
PixelCNN, Wavenet, Normalizing Flows - Santiago Pascual - UPC Barcelona 2018
 
Markov Chain Monitoring - Application to demand prediction in bike sharing sy...
Markov Chain Monitoring - Application to demand prediction in bike sharing sy...Markov Chain Monitoring - Application to demand prediction in bike sharing sy...
Markov Chain Monitoring - Application to demand prediction in bike sharing sy...
 
Linear Size Meshes
Linear Size MeshesLinear Size Meshes
Linear Size Meshes
 
エンドツーエンド音声合成に向けたNIIにおけるソフトウェア群 ~ TacotronとWaveNetのチュートリアル (Part 1)~
エンドツーエンド音声合成に向けたNIIにおけるソフトウェア群 ~ TacotronとWaveNetのチュートリアル (Part 1)~エンドツーエンド音声合成に向けたNIIにおけるソフトウェア群 ~ TacotronとWaveNetのチュートリアル (Part 1)~
エンドツーエンド音声合成に向けたNIIにおけるソフトウェア群 ~ TacotronとWaveNetのチュートリアル (Part 1)~
 
The Power of Motif Counting Theory, Algorithms, and Applications for Large Gr...
The Power of Motif Counting Theory, Algorithms, and Applications for Large Gr...The Power of Motif Counting Theory, Algorithms, and Applications for Large Gr...
The Power of Motif Counting Theory, Algorithms, and Applications for Large Gr...
 
Convolutional Neural Networks - Veronica Vilaplana - UPC Barcelona 2018
Convolutional Neural Networks - Veronica Vilaplana - UPC Barcelona 2018Convolutional Neural Networks - Veronica Vilaplana - UPC Barcelona 2018
Convolutional Neural Networks - Veronica Vilaplana - UPC Barcelona 2018
 
Queuing theory and traffic analysis in depth
Queuing theory and traffic analysis in depthQueuing theory and traffic analysis in depth
Queuing theory and traffic analysis in depth
 
Lecture 02 internet video search
Lecture 02 internet video searchLecture 02 internet video search
Lecture 02 internet video search
 

Kürzlich hochgeladen

UGC NET Paper 1 Mathematical Reasoning & Aptitude.pdf
UGC NET Paper 1 Mathematical Reasoning & Aptitude.pdfUGC NET Paper 1 Mathematical Reasoning & Aptitude.pdf
UGC NET Paper 1 Mathematical Reasoning & Aptitude.pdfNirmal Dwivedi
 
Micro-Scholarship, What it is, How can it help me.pdf
Micro-Scholarship, What it is, How can it help me.pdfMicro-Scholarship, What it is, How can it help me.pdf
Micro-Scholarship, What it is, How can it help me.pdfPoh-Sun Goh
 
Key note speaker Neum_Admir Softic_ENG.pdf
Key note speaker Neum_Admir Softic_ENG.pdfKey note speaker Neum_Admir Softic_ENG.pdf
Key note speaker Neum_Admir Softic_ENG.pdfAdmir Softic
 
microwave assisted reaction. General introduction
microwave assisted reaction. General introductionmicrowave assisted reaction. General introduction
microwave assisted reaction. General introductionMaksud Ahmed
 
1029 - Danh muc Sach Giao Khoa 10 . pdf
1029 -  Danh muc Sach Giao Khoa 10 . pdf1029 -  Danh muc Sach Giao Khoa 10 . pdf
1029 - Danh muc Sach Giao Khoa 10 . pdfQucHHunhnh
 
Dyslexia AI Workshop for Slideshare.pptx
Dyslexia AI Workshop for Slideshare.pptxDyslexia AI Workshop for Slideshare.pptx
Dyslexia AI Workshop for Slideshare.pptxcallscotland1987
 
Unit-V; Pricing (Pharma Marketing Management).pptx
Unit-V; Pricing (Pharma Marketing Management).pptxUnit-V; Pricing (Pharma Marketing Management).pptx
Unit-V; Pricing (Pharma Marketing Management).pptxVishalSingh1417
 
How to Manage Global Discount in Odoo 17 POS
How to Manage Global Discount in Odoo 17 POSHow to Manage Global Discount in Odoo 17 POS
How to Manage Global Discount in Odoo 17 POSCeline George
 
Application orientated numerical on hev.ppt
Application orientated numerical on hev.pptApplication orientated numerical on hev.ppt
Application orientated numerical on hev.pptRamjanShidvankar
 
Activity 01 - Artificial Culture (1).pdf
Activity 01 - Artificial Culture (1).pdfActivity 01 - Artificial Culture (1).pdf
Activity 01 - Artificial Culture (1).pdfciinovamais
 
How to Give a Domain for a Field in Odoo 17
How to Give a Domain for a Field in Odoo 17How to Give a Domain for a Field in Odoo 17
How to Give a Domain for a Field in Odoo 17Celine George
 
SKILL OF INTRODUCING THE LESSON MICRO SKILLS.pptx
SKILL OF INTRODUCING THE LESSON MICRO SKILLS.pptxSKILL OF INTRODUCING THE LESSON MICRO SKILLS.pptx
SKILL OF INTRODUCING THE LESSON MICRO SKILLS.pptxAmanpreet Kaur
 
The basics of sentences session 3pptx.pptx
The basics of sentences session 3pptx.pptxThe basics of sentences session 3pptx.pptx
The basics of sentences session 3pptx.pptxheathfieldcps1
 
Food safety_Challenges food safety laboratories_.pdf
Food safety_Challenges food safety laboratories_.pdfFood safety_Challenges food safety laboratories_.pdf
Food safety_Challenges food safety laboratories_.pdfSherif Taha
 
Holdier Curriculum Vitae (April 2024).pdf
Holdier Curriculum Vitae (April 2024).pdfHoldier Curriculum Vitae (April 2024).pdf
Holdier Curriculum Vitae (April 2024).pdfagholdier
 
Unit-IV- Pharma. Marketing Channels.pptx
Unit-IV- Pharma. Marketing Channels.pptxUnit-IV- Pharma. Marketing Channels.pptx
Unit-IV- Pharma. Marketing Channels.pptxVishalSingh1417
 
ComPTIA Overview | Comptia Security+ Book SY0-701
ComPTIA Overview | Comptia Security+ Book SY0-701ComPTIA Overview | Comptia Security+ Book SY0-701
ComPTIA Overview | Comptia Security+ Book SY0-701bronxfugly43
 
psychiatric nursing HISTORY COLLECTION .docx
psychiatric  nursing HISTORY  COLLECTION  .docxpsychiatric  nursing HISTORY  COLLECTION  .docx
psychiatric nursing HISTORY COLLECTION .docxPoojaSen20
 

Kürzlich hochgeladen (20)

UGC NET Paper 1 Mathematical Reasoning & Aptitude.pdf
UGC NET Paper 1 Mathematical Reasoning & Aptitude.pdfUGC NET Paper 1 Mathematical Reasoning & Aptitude.pdf
UGC NET Paper 1 Mathematical Reasoning & Aptitude.pdf
 
Spatium Project Simulation student brief
Spatium Project Simulation student briefSpatium Project Simulation student brief
Spatium Project Simulation student brief
 
Micro-Scholarship, What it is, How can it help me.pdf
Micro-Scholarship, What it is, How can it help me.pdfMicro-Scholarship, What it is, How can it help me.pdf
Micro-Scholarship, What it is, How can it help me.pdf
 
Key note speaker Neum_Admir Softic_ENG.pdf
Key note speaker Neum_Admir Softic_ENG.pdfKey note speaker Neum_Admir Softic_ENG.pdf
Key note speaker Neum_Admir Softic_ENG.pdf
 
microwave assisted reaction. General introduction
microwave assisted reaction. General introductionmicrowave assisted reaction. General introduction
microwave assisted reaction. General introduction
 
1029 - Danh muc Sach Giao Khoa 10 . pdf
1029 -  Danh muc Sach Giao Khoa 10 . pdf1029 -  Danh muc Sach Giao Khoa 10 . pdf
1029 - Danh muc Sach Giao Khoa 10 . pdf
 
Dyslexia AI Workshop for Slideshare.pptx
Dyslexia AI Workshop for Slideshare.pptxDyslexia AI Workshop for Slideshare.pptx
Dyslexia AI Workshop for Slideshare.pptx
 
Unit-V; Pricing (Pharma Marketing Management).pptx
Unit-V; Pricing (Pharma Marketing Management).pptxUnit-V; Pricing (Pharma Marketing Management).pptx
Unit-V; Pricing (Pharma Marketing Management).pptx
 
How to Manage Global Discount in Odoo 17 POS
How to Manage Global Discount in Odoo 17 POSHow to Manage Global Discount in Odoo 17 POS
How to Manage Global Discount in Odoo 17 POS
 
Application orientated numerical on hev.ppt
Application orientated numerical on hev.pptApplication orientated numerical on hev.ppt
Application orientated numerical on hev.ppt
 
Activity 01 - Artificial Culture (1).pdf
Activity 01 - Artificial Culture (1).pdfActivity 01 - Artificial Culture (1).pdf
Activity 01 - Artificial Culture (1).pdf
 
How to Give a Domain for a Field in Odoo 17
How to Give a Domain for a Field in Odoo 17How to Give a Domain for a Field in Odoo 17
How to Give a Domain for a Field in Odoo 17
 
SKILL OF INTRODUCING THE LESSON MICRO SKILLS.pptx
SKILL OF INTRODUCING THE LESSON MICRO SKILLS.pptxSKILL OF INTRODUCING THE LESSON MICRO SKILLS.pptx
SKILL OF INTRODUCING THE LESSON MICRO SKILLS.pptx
 
The basics of sentences session 3pptx.pptx
The basics of sentences session 3pptx.pptxThe basics of sentences session 3pptx.pptx
The basics of sentences session 3pptx.pptx
 
Food safety_Challenges food safety laboratories_.pdf
Food safety_Challenges food safety laboratories_.pdfFood safety_Challenges food safety laboratories_.pdf
Food safety_Challenges food safety laboratories_.pdf
 
Mehran University Newsletter Vol-X, Issue-I, 2024
Mehran University Newsletter Vol-X, Issue-I, 2024Mehran University Newsletter Vol-X, Issue-I, 2024
Mehran University Newsletter Vol-X, Issue-I, 2024
 
Holdier Curriculum Vitae (April 2024).pdf
Holdier Curriculum Vitae (April 2024).pdfHoldier Curriculum Vitae (April 2024).pdf
Holdier Curriculum Vitae (April 2024).pdf
 
Unit-IV- Pharma. Marketing Channels.pptx
Unit-IV- Pharma. Marketing Channels.pptxUnit-IV- Pharma. Marketing Channels.pptx
Unit-IV- Pharma. Marketing Channels.pptx
 
ComPTIA Overview | Comptia Security+ Book SY0-701
ComPTIA Overview | Comptia Security+ Book SY0-701ComPTIA Overview | Comptia Security+ Book SY0-701
ComPTIA Overview | Comptia Security+ Book SY0-701
 
psychiatric nursing HISTORY COLLECTION .docx
psychiatric  nursing HISTORY  COLLECTION  .docxpsychiatric  nursing HISTORY  COLLECTION  .docx
psychiatric nursing HISTORY COLLECTION .docx
 

Scalable membership management

  • 1. Scalable membership management and failure detection? Vinay Setty INF5360
  • 2. What is Gossiping? • Spread of information in a random manner • Some examples: – Human gossiping – Epidemic diseases – Physical phenomenon: wild fire, diffusion etc – Computer viruses and worms
  • 3. Gossiping in Computer Science • Term first coined by Demers et al (1987) • Some applications of gossip protocols – Peer Sampling – Data Aggregation – Clustering – Information Dissemination (Multicast, Pub/Sub) – Overlay/topology – Maintenance – Failure detection?
  • 4. Gossip-Based Protocol: Example 3 4 0 5 1 2 9 8 7 6
  • 5. Today’s Focus • Theoretical angle for Gossip-based protocols [Allavena et al PODC 2005] – Probability of partitioning – Time till partitioning – Bounds on in-degree – Essential elements of gossiping – Simulation results • Cyclon [Voulgaris et al] • Scamp [Ganesh et al] • NewsCast [Jelasity et al]
  • 6. Membership Service • Full Membership – Complete knowledge at each node – Random subset used for gossiping – Not scalable – Hard to maintain • Partial Membership – Random subset at each node – Gossip partners chosen from local view
  • 7. View Selection L1 s s,p,r p v s,p,t u r t t,q,r q L2 L1 L2 v Weighted with w
  • 8. Essential Elements of Gossiping • Mixing: Construct a list L1 consisting of local views of local view of node u – Guarantees non partitioning – “Pull” based • Reinforcement: Construct a list L2 consisting of local views of nodes that requested local view of u – Balances network – removes old possibly dead edges, adds new edges
  • 9. Partitioning and Size Estimate • A and B partition iff x=1 and y=0 • Partitioning is least possible when x=y • Goal of protocol is to maintain this balance
  • 10. Size Estimates • Idea: – Assuming edges were drawn uniformly randomly, expected x+y  |A| – x is estimate of size of A by nodes in A – y is estimate of size of A by nodes in B • Mixing: – Agreeing on estimation of x and y ensures no partition (even if x and y are not accurate) • Reinforcement: – Brings estimation of x and y to correct value
  • 11. K-regularity • View Size: k • Number of nodes: n • Fraction of nodes in partition: γ • |A|= γn ≤ |B| • #edges from A to B: (1-x)γkn • #edges from B to A: y (1-γ)kn • Number of edges in A-B cut: – (1-x)γkn +x (1-γ)kn (since x=y) – ≥ γkn (assuming γ≤½)
  • 12. Time Till Partitioning • View Size: k • Number of nodes: n • Fraction of nodes in partition: γ • Churn rate: μ (μn nodes leave and join) • Claim: Expected time before a partition of size γ happens ≈ 2γkn – As long as μ≪γkn
  • 13. 100, 000 nodes, view sizes of 17, a fanout of 3, and a loosely synchronised syst em, t he maximum in-degree was always which re-samples ran- below 4.5 t imes t hat of a random graph and t he st andard g t he names of t he nodes deviat ion was not more t han 3.2 t imes larger t han t hat of a Iterations until Partitioning y or anot her is doomed enat ion of all t he views. random graph. T hese values would improve wit h increased fanout , but even a fanout of 2 gives sat isfact ory perfor- onds t o creat ing a new mance. eplacement from t he old om V at each it erat ion. 10000 annot reappear wit hout cement . T he diversit y 9000 ime, and in fact rat her 8000 Number of iterations rk. Not e t hat it is t he- 7000 e by creat ing a protnodes: n Number of ocol 6000 t at ion on V, size: tk = log n View but his is Churn: n/32 esn’t necessarily behave 5000 ving or joining t he net - 4000 ively add t he names of 3000 t o V, a process we call ome reinforcement , even 2000 n t he art icle: each pro- 1000 hen sending a message. 0 nd Cyclon [16] as well: 1 1.5 2 2.5 3 3.5 eir view t hat t hey t hen Log10 of the number of nodes aviour in say t he cont ext F igure 4: N umber of it erat ions unt il part it ioning t he “ News Event s” are des. Let only t he nodes We were int erest ed in mat ching our t heoret ical result s odes add t heir names t o about part it ioning and churn. We ran simulat ions evaluat - Event s” inst ead of every ing t he number of it erat ions unt il part it ioning. By part it ion-
  • 14. View Size vs Time until Partition Number of nodes: n View size: k = log n Churn: n/32
  • 15. Simplified Model for Proof – Single randomly chosen element from view is replaced instead of whole views – Assumption: The out-edges of nodes of A are identically distributed and same applies to B – a = #edges from A to A – c = #edges from A to B – b = #edges from B to A – d = #edges from B to B
  • 16. Proof Intuition Partition state: a = γkn and b = 0
  • 17. In-Degree Analysis • Load balancing requires balance in in-degree distribution • In-degree is governed by the way edges created, copied and destroyed • Copying some edges more than others cause variability in in-degree • Node living longer is expected to have higher in- degree • Solution: Increase reinforcement and keep track of timestamps like in Cyclon • Simulation: max in-degree < 4.5 times of random graph and standard deviation < 3.2 times
  • 18. Discussion • Are these theoretical guarantees practically useful? • Goal is not provide failure detection
  • 19. Cyclon • Consists of same elements as suggested by [Allavena et al PODC 2005] • [Allavena et al PODC 2005] Analysis holds for Cyclon • Major differences: – Timestamps – shuffling
  • 20. Basic Shuffling • Select a random subset of l neighbors (1 ≤ l ≤ c) from P’s own cache, and a random peer, Q, within this subset, where l is a system parameter, called shuffle length. • Replace Q’s address with P’s address. • Send the updated subset to Q. • Receive from Q a subset of no more than l of Q’s neighbors. • Discard entries pointing to P, and entries that are already in P’s cache. • Update P’s cache to include all remaining entries, by – firstly using empty cache slots (if any), and – secondly replacing entries among the ones originally sent to Q.
  • 22. Enhanced Shuffling • Increase by one the age of all neighbors. • Select neighbor Q with the highest age among all neighbors, and l − 1 other random neighbors. • Replace Q’s entry with a new entry of age 0 and with P’s address. • Send the updated subset to peer Q. • Receive from Q a subset of no more that l of its own entries. • Discard entries pointing at P and entries already contained in P’s cache. • Update P’s cache to include all remaining entries, by firstly using empty • cache slots (if any), and secondly replacing entries among the ones sent to Q.
  • 23. Time Until Dead Links Removed
  • 24. removed. Note that the number of clusters decreases as we approach 100% node removal because the total number of surviving nodes becomes too small. Fig- Number of Clusters ure 7(b) shows the number of nodes not belonging to the largest cluster, in log scale. These graphs show considerable robustness to node failures, especially con- sidering the fact that in the early stages of clustering very few nodes are out of the largest cluster, which indicates that most nodes are still connected in a single Fig. 7. (a) Number of disjoint clusters, as a result of removing a large percentage of nodes. Shows that the overlay does not break into two or more disjoint clusters, unless a major percentage of the nodes are removed. (b) Number of nodes not belonging to the largest cluster. Shows that in the first steps of clustering only a few nodes are separated from the main cluster, which still connects the
  • 27. SCAMP • Partial knowledge of the membership: local view • Fanout automatically set = size of the local view • Fanout evolves naturally with the size of the group – Size of local views converges towards C.log(n)
  • 28. Join (Subscription) Subscription to Subscription forwarded a random member P=1/sizeof view s 1 s S 0 (1-P) s s P=1/sizeof view s 2 s (1-P) s P=1/sizeof view 3 s (1-P)
  • 29. Join(Subscription) algorithm 7 6 Local view 1 4 5 6 4 0 6 7 2 3 6 6 6 1 2 6 0 8 3 6 7 0 1 5 6 6 6 5 8 7
  • 30. Load Balancing • Indirection: – Forward the subscription instead of handling request • Lease associated with each subscription • Periodically nodes have to re-subscribe – Nodes having failed permanently will time out – Re-balance the partial views
  • 31. Unsubscription Local view 8 9 0 8 9 4 1 4 5 Unsub (0), [1,4,5] 4 x x 0 7 3 0 7 3 5 1 y y 6 0 2 6 0 1 5 z z
  • 32. Degree • System modelled as random directed graph • D(N) = Average out-degree for N-nodes system • Subscription adds D(N)+1 directed arcs, so • (N+1) D(N+1) = N D(N) + D(N)+1 • Solution of this recursion is • D(N)=D(1)+1/2+1/3+…+1/N  Log(N)
  • 33. Distribution of view size 35000 30000 Log=13.12 25000 Number of nodes 20000 15000 10000 Log=12.2 5000 0 0 5 10 15 20 25 30 35 40 45 50 View Size 200 000 Node System 500 000 Node System 33
  • 34. Reliability: 5000 node system 1 0.98 0.96 Reliability 0.94 0.92 0.9 0 500 1000 1500 2000 2500 Number of failures SCAMP Global membership knowledge, fanout=8 Global membership knowledge, fanout=9 34
  • 35. NewsCast • Goal: Aggregate information in – a large and dynamic – distributed environment – a robust and dependable manner
  • 36. Idea • Gets news from application, timestamps it and adds local peer address to the cache entry • Finds a random peer in cache addresses – Sends all cache entries to this peer – Receives all cache entries from that peer • Passes on cache entries (containing news items) to application • Merges old cache with received cache – Keeps at most C most recent cache entries
  • 37. Aggregation • Each node ni maintains a single number xi • Every node ni selects a random node nk, and sends its value xi to nk • nk responds with the aggregate (e.g. max(xi,xk) ) of the incoming and its own value • 4. Aggregate values will converge “exponentially”
  • 38. Path length under failures
  • 40. Aggregation 1 proportion of not-reached nodes 0.1 0.01 0.001 theoretical model c=20 c=40 c=80 0.0001 6 7 8 9 10 11 cycle

Hinweis der Redaktion

  1. (a)Timeuntildeadnodesareforgotten.(b)Numberofdeadlinks.
  2. Fig.7. (a)Numberofdisjointclusters,asaresultofremovingalargepercentageofnodes.Shows that the overlay does not break into two or more disjoint clusters, unless a major percentage of the nodes are removed. (b) Number of nodes not belonging to the largest cluster. Shows that in the first steps of clustering only a few nodes are separated from the main cluster, which still connects the grand majority of the nodes.
  3. Note that the graph for the experiment with cache size 100 is practically a flat line. That is, for 100,000 nodes and cache size 100, the overlay created is so robust, that no matter how many nodes are removed, the remaining ones remain connected in a single cluster.
  4. In-degreedistributioninconverged100,000nodeoverlay,forbasicshuffling,enhanced shuffling, and an overlay where each node has c randomly chosen outgoing links. t is, however, clear that enhanced shuffling does a significantly better job with respect to spreading out the links extremely evenly across all nodes. For the experiment with cache size 20, 80.31% of the nodes have an in-degree of 20 ± 5%. For the experiment with cache size 50, 93.95% of the nodes have an in-degree of 50 ± 5%. The respective percentages for basic shuffling are 36.22 and 38.47%.
  5. This is an average case analysis. In reality, there are noise terms in this recurrence as we pick a node whose degree is only approximately d(N). In order to prove that the argument is correct in the presence of this noise, we need to control the variance of that noise (and invoke the martingale convergence theorem!).
  6. to the overall value incase the aggregation is a average function, and “super- exponentially” incase of a maximum function
  7. maximum finding protocol. N = 105 , points are averages of 50 runs. Standard deviation is not shown, it is several orders of magnitude lower than the average.