CORRECTNESS OF A GOSSIP BASED MEMBERSHIP PROTOCOL BY (ANDRÉ ALLAVENA, ALAN DEMERS, JOHN E. HOPCROFT ) PRATIK TIMALSENA UNIVERSITY OF OSLO

Size: px
Start display at page:

Download "CORRECTNESS OF A GOSSIP BASED MEMBERSHIP PROTOCOL BY (ANDRÉ ALLAVENA, ALAN DEMERS, JOHN E. HOPCROFT ) PRATIK TIMALSENA UNIVERSITY OF OSLO"

Transcription

1 CORRECTNESS OF A GOSSIP BASED MEMBERSHIP PROTOCOL BY (ANDRÉ ALLAVENA, ALAN DEMERS, JOHN E. HOPCROFT ) PRATIK TIMALSENA UNIVERSITY OF OSLO

2 OUTLINE q Contribution of the paper q Gossip algorithm q The corrected Gossip algorithm q Maintenance of the local view q Model and definition q Simulation and result q Summary and discussion 2

3 Contribution of the paper q scalable gossip-based algorithm for local view maintenance. q Can be combined with any application level gossip protocol that relies on randomly selected gossip partners q Preserve connectivity and load balancing between nodes 3

4 GOSSIP ALGORITHM q Practical example when gossip algorithm required q catering example q Catering people stock on snow storm q Party without food is boring q How to pass message to all to bring food for party q One option calling tree 4

5 GOSSIP ALGORITHM 5

6 GOSSIP ALGORITHM q Gossip based membership q Motivation Importance of scalability and fault-tolerance in distributed system Has led to considerable research in multicast protocols using gossip q Each node forwards a message to set of gossip targets q Probabilistic guarantees of delivery. q Reliability By setting amount of gossip targets large 6

7 GOSSIP ALGORITHM q Generic protocol q At every round, each node u: q selects at random f nodes from its current view; q contacts these f nodes and requests their views; q concatenates these views into a list L; q adds to its list L the nodes that request u s view during the round, if any; q adds to its list L the views of the nodes that request u s view during the round, if any; q creates its new view by selecting k elements from the list L, not allowing duplicates. 7

8 GOSSIP ALGORITHM 8

9 GOSSIP ALGORITHM q Generic protocol q Potential issue q The network could break into two or more disconnected components. no links between components; q The network could become unbalanced,( swamping a few nodes with a very high load instead of spreading the load more or less evenly) q The views could not mix well, and the system at time t = O(ln n) would not be independent from the system at time t = 0. 9

10 GOSSIP ALGORITHM q Variant gossip algorithm(corrected one) q Desirable properties q Load balancing(reinforcement, probabilistic bound on degree of node) q Scalability (local view) q Avoiding portioning(even distribution of the pointer to the nodes) 10

11 GOSSIP ALGORITHM CORRECTED q How it works q Protocol is based on each node having a local view l Fixed-size random subset of group membership q N number of nodes q K the size of a local view q F fanout parameter q W reinforcement weight q local view of random node 11

12 GOSSIP ALGORITHM CORRECTED 12

13 GOSSIP ALGORITHM CORRECTION q construct a list L1 comprising the local views of f nodes chosen at random from the local view of s q construct a list L2 of the other nodes that requested its view during the round q create a new local view by choosing k distinct elements at random from L1 andl2 q Reinforcement weight ω determines how much more likely nodes are to be selected from L2 than L1. q If ω is 0, nodes in L2 are ignored q if ω is 1, we make no distinction between L1 and L2 q In the limit ω goes to, all nodes are taken from L2 if possible.

14 GOSSIP ALGORITHM CORRECTION q Join and leave q Node join by copying the local view of random node q Leave by stopping to participate leaving the dangled age in the graph q No use of keep alive and heartbeat message q No difference between stopping and failing

15 Explanation of the functioning of the protocol q The view requested by u q The node that requested u s view q Mixing, pulling the view from several nodes q The node u pulls the node that request its view in the list where it creates the new view q Adding its name to the list(pool of names) rather than information of view q Provides the balance q Without reinforcement network collapse like star graph 15

16 PARTITIONING AND ESTIMATE SIZE q Consider partition set A and B q x fraction edge from A nodes to A q y fraction edge from B to A q γ = A /n fraction of A nodes q If edges are drawn normally Then the edges from A or B to A is proportional to normal size Of A q x =30% y=70% set S q x =y q x-y = 1 16

17 EVOLUTION OF ESTIMATE OF SIZE OF A q Mixing q Act of pulling and merging the views corresponding to checking nodes from both sides for their estimate and averaging them q Converges fast at x = y q x=0, y=1 q Provides graphs not partitioned q Provides necessary condition for reinforcement to drag to correct value of the size 17

18 EVOLUTION OF ESTIMATE OF SIZE OF A q Reinforcement q Both side agreeing on their estimate of size of A q Some node from A and some from B are going to pull A nodes q At what proportion? q Same proportion as the size of A and B q x=y=30% q A nodes pull 30% from A and so does B nodes q A sees 30% of A nodes and B as well q Removes older/dead edges in approximately K/F rounds q adds fresh edges q Thus brings the estimate of size of A to correct value q Adds a little bit to estimation at each iteration leads to correct value

19 MODEL AND DEFINITION q Non partitioning q Expected time before a fraction γ of the nodes partiton away from the rest of the nodes is in exponential in γkn. q k is size of the view q n is number of nodes in the network q Above facts says that only small component can be disconnected and rest stays connected q Disconnected can easily detected by observing diversity in the content value over time q And attempted to rejoin q View size needs to be larger than churn rate for expected time until partition to be exponential

20 MODEL AND DEFINITION q Model intuition q Each iteration node v q Each node has some number of edges pointing to nodes in A, and some number of edges pointing to nodes in B q a node s edge distribution to be independent of that of its neighbors. q In u v A is like x A

21 MODEL AND DEFINITION q Model q If the element being updated is a 0, the new value is the value of a position chosen uniformly at random from A. q If it is a 1, the value is the value of a position chosen uniformly at random from B. q for reinforcement, if the node doing the tagging is in A, put a zero, otherwise, a one.

22 MODEL AND DEFINITION

23 MODEL AND DEFINITION q Limit probability distribution q The limit probability distribution is 1 for the partitioned state (a = γkn, b = 0), and 0 for the rest of the grid q add a transition out of the partitioned state to the two neighbor states(a=γkn 1,b=0 ) and(a=γkn,b=1). q Claim q the fraction of time spent in the partitioned state of this finite (2-dimensional) Markov Chain is exponentially small in γkn when µ γkn and p 1/ ln γkn

24 Simplified Model q Array Υ of length kn (concatenation of the view) q Assumed no nodes leaves or join q At each iteration randomly chosen element replaced by q Probability p picked node uniformly at random q With probability 1-p copy value of the element chosen uniformly at random q When reinforcing a node replaces elements in view by a node randomly chosen from list of nodes in the system q When pulling by arbitrary view. 24

25 Simplified Model q No difference between picking from a view from A or B( distribution governing the edge distribution is the same for all nodes ) q Justified when agreed on the partition size estimate q the conditioning on neighbors to be negligible 25

26 Simplified Model q Without reinforcement q P=0 in simplified model q Without reinforcement nodes disappear quickly from union each round q Cannot bring back without reinforcement q The diversity of content of the Υ decreases and the network collapse into star like graph 26

27 Simplified Model q With reinforcement q Can solve for the limit distribution q Find correct value for the estimate of the size of the partition. q Analysis shows both side once agreed on estimate of the size of the partition q Estimate converge to correct value and stay there(network balance) 27

28 Simulation Results q Simulations show that performance of protocol is good q For large scale test with up to nodes, view size 17, fanout 3 q Almost as good as a random graph q Maximum in-degree always below 4.5 times that of random graph q High number of rounds before any partitioning happens q Scales well with increasing number of nodes 28

29 Simulation Results 29

30 Alternative protocol q Several choices made in the design of the algorithm: push, pull, randomize etc. q Explore some alternative and cite alternative work q Doesn t distinguish between pull and reinforcement. q Reinforcement q Used push method q Add actively one s name to the a (multi)set defined by concatenation view(υ) q To compensate for the drop due to randomness in the pulling part or node failure q Alternative pulling method q Adding node u to one s view is useful if not present already q The nodes which cannot contact u can be the case 30

31 Alternative protocol q Mixing q Used pull mechanism q Mixing of the views q Push pull and push-pull are alternatives q Pull is better according to author q Random walking on the directed graph leads to the view of the node (2^t) q In push random walk not directed and can be back tracked and increases the risk of portioning q Randomize q Instead of making all choices random, use timestamp for every view entry q Randomly select view on the the timestamp q Old edge replaced by fresh edges q CYCLON similar but uses the timestamp 31

32 SUMMARY AND DISCUSSION q Satisfactory protocol for local view maintenance q Provide the way to preserve connectivity q Provides the network balance q Only simulation is done and the real implementation 32

Distributed Systems Gossip Algorithms

Distributed Systems Gossip Algorithms Distributed Systems Gossip Algorithms He Sun School of Informatics University of Edinburgh What is Gossip? Gossip algorithms In a gossip algorithm, each node in the network periodically exchanges information

More information

Coupling. 2/3/2010 and 2/5/2010

Coupling. 2/3/2010 and 2/5/2010 Coupling 2/3/2010 and 2/5/2010 1 Introduction Consider the move to middle shuffle where a card from the top is placed uniformly at random at a position in the deck. It is easy to see that this Markov Chain

More information

Propp-Wilson Algorithm (and sampling the Ising model)

Propp-Wilson Algorithm (and sampling the Ising model) Propp-Wilson Algorithm (and sampling the Ising model) Danny Leshem, Nov 2009 References: Haggstrom, O. (2002) Finite Markov Chains and Algorithmic Applications, ch. 10-11 Propp, J. & Wilson, D. (1996)

More information

Coordination. Failures and Consensus. Consensus. Consensus. Overview. Properties for Correct Consensus. Variant I: Consensus (C) P 1. v 1.

Coordination. Failures and Consensus. Consensus. Consensus. Overview. Properties for Correct Consensus. Variant I: Consensus (C) P 1. v 1. Coordination Failures and Consensus If the solution to availability and scalability is to decentralize and replicate functions and data, how do we coordinate the nodes? data consistency update propagation

More information

Dominating Set. Chapter 7

Dominating Set. Chapter 7 Chapter 7 Dominating Set In this chapter we present another randomized algorithm that demonstrates the power of randomization to break symmetries. We study the problem of finding a small dominating set

More information

Graph-theoretic Problems

Graph-theoretic Problems Graph-theoretic Problems Parallel algorithms for fundamental graph-theoretic problems: We already used a parallelization of dynamic programming to solve the all-pairs-shortest-path problem. Here we are

More information

6.842 Randomness and Computation March 3, Lecture 8

6.842 Randomness and Computation March 3, Lecture 8 6.84 Randomness and Computation March 3, 04 Lecture 8 Lecturer: Ronitt Rubinfeld Scribe: Daniel Grier Useful Linear Algebra Let v = (v, v,..., v n ) be a non-zero n-dimensional row vector and P an n n

More information

Lecture December 2009 Fall 2009 Scribe: R. Ring In this lecture we will talk about

Lecture December 2009 Fall 2009 Scribe: R. Ring In this lecture we will talk about 0368.4170: Cryptography and Game Theory Ran Canetti and Alon Rosen Lecture 7 02 December 2009 Fall 2009 Scribe: R. Ring In this lecture we will talk about Two-Player zero-sum games (min-max theorem) Mixed

More information

4 : Exact Inference: Variable Elimination

4 : Exact Inference: Variable Elimination 10-708: Probabilistic Graphical Models 10-708, Spring 2014 4 : Exact Inference: Variable Elimination Lecturer: Eric P. ing Scribes: Soumya Batra, Pradeep Dasigi, Manzil Zaheer 1 Probabilistic Inference

More information

c 2010 Society for Industrial and Applied Mathematics

c 2010 Society for Industrial and Applied Mathematics SIAM J. COMPUT. Vol. 39, No. 8, pp. 3830 3859 c 2010 Society for Industrial and Applied Mathematics CORRECTNESS OF GOSSIP-BASED MEMBERSHIP UNDER MESSAGE LOSS MAXIM GUREVICH AND IDIT KEIDAR Abstract. Due

More information

Distributed Systems Byzantine Agreement

Distributed Systems Byzantine Agreement Distributed Systems Byzantine Agreement He Sun School of Informatics University of Edinburgh Outline Finish EIG algorithm for Byzantine agreement. Number-of-processors lower bound for Byzantine agreement.

More information

CPSC 467: Cryptography and Computer Security

CPSC 467: Cryptography and Computer Security CPSC 467: Cryptography and Computer Security Michael J. Fischer Lecture 14 October 16, 2013 CPSC 467, Lecture 14 1/45 Message Digest / Cryptographic Hash Functions Hash Function Constructions Extending

More information

arxiv: v1 [cs.dc] 22 Oct 2018

arxiv: v1 [cs.dc] 22 Oct 2018 FANTOM: A SCALABLE FRAMEWORK FOR ASYNCHRONOUS DISTRIBUTED SYSTEMS A PREPRINT Sang-Min Choi, Jiho Park, Quan Nguyen, and Andre Cronje arxiv:1810.10360v1 [cs.dc] 22 Oct 2018 FANTOM Lab FANTOM Foundation

More information

Amortized Complexity Main Idea

Amortized Complexity Main Idea omp2711 S1 2006 Amortized Complexity Example 1 Amortized Complexity Main Idea Worst case analysis of run time complexity is often too pessimistic. Average case analysis may be difficult because (i) it

More information

6.207/14.15: Networks Lecture 7: Search on Networks: Navigation and Web Search

6.207/14.15: Networks Lecture 7: Search on Networks: Navigation and Web Search 6.207/14.15: Networks Lecture 7: Search on Networks: Navigation and Web Search Daron Acemoglu and Asu Ozdaglar MIT September 30, 2009 1 Networks: Lecture 7 Outline Navigation (or decentralized search)

More information

Fault-Tolerant Consensus

Fault-Tolerant Consensus Fault-Tolerant Consensus CS556 - Panagiota Fatourou 1 Assumptions Consensus Denote by f the maximum number of processes that may fail. We call the system f-resilient Description of the Problem Each process

More information

Lecture 2: Divide and conquer and Dynamic programming

Lecture 2: Divide and conquer and Dynamic programming Chapter 2 Lecture 2: Divide and conquer and Dynamic programming 2.1 Divide and Conquer Idea: - divide the problem into subproblems in linear time - solve subproblems recursively - combine the results in

More information

Distributed Systems Principles and Paradigms. Chapter 06: Synchronization

Distributed Systems Principles and Paradigms. Chapter 06: Synchronization Distributed Systems Principles and Paradigms Maarten van Steen VU Amsterdam, Dept. Computer Science Room R4.20, steen@cs.vu.nl Chapter 06: Synchronization Version: November 16, 2009 2 / 39 Contents Chapter

More information

We are going to discuss what it means for a sequence to converge in three stages: First, we define what it means for a sequence to converge to zero

We are going to discuss what it means for a sequence to converge in three stages: First, we define what it means for a sequence to converge to zero Chapter Limits of Sequences Calculus Student: lim s n = 0 means the s n are getting closer and closer to zero but never gets there. Instructor: ARGHHHHH! Exercise. Think of a better response for the instructor.

More information

Dominating Set. Chapter 26

Dominating Set. Chapter 26 Chapter 26 Dominating Set In this chapter we present another randomized algorithm that demonstrates the power of randomization to break symmetries. We study the problem of finding a small dominating set

More information

Dominating Set. Chapter Sequential Greedy Algorithm 294 CHAPTER 26. DOMINATING SET

Dominating Set. Chapter Sequential Greedy Algorithm 294 CHAPTER 26. DOMINATING SET 294 CHAPTER 26. DOMINATING SET 26.1 Sequential Greedy Algorithm Chapter 26 Dominating Set Intuitively, to end up with a small dominating set S, nodes in S need to cover as many neighbors as possible. It

More information

CS5314 Randomized Algorithms. Lecture 15: Balls, Bins, Random Graphs (Hashing)

CS5314 Randomized Algorithms. Lecture 15: Balls, Bins, Random Graphs (Hashing) CS5314 Randomized Algorithms Lecture 15: Balls, Bins, Random Graphs (Hashing) 1 Objectives Study various hashing schemes Apply balls-and-bins model to analyze their performances 2 Chain Hashing Suppose

More information

Distributed Systems Principles and Paradigms

Distributed Systems Principles and Paradigms Distributed Systems Principles and Paradigms Chapter 6 (version April 7, 28) Maarten van Steen Vrije Universiteit Amsterdam, Faculty of Science Dept. Mathematics and Computer Science Room R4.2. Tel: (2)

More information

Lecture 8: Decision-making under total uncertainty: the multiplicative weight algorithm. Lecturer: Sanjeev Arora

Lecture 8: Decision-making under total uncertainty: the multiplicative weight algorithm. Lecturer: Sanjeev Arora princeton univ. F 13 cos 521: Advanced Algorithm Design Lecture 8: Decision-making under total uncertainty: the multiplicative weight algorithm Lecturer: Sanjeev Arora Scribe: (Today s notes below are

More information

CPSC 467: Cryptography and Computer Security

CPSC 467: Cryptography and Computer Security CPSC 467: Cryptography and Computer Security Michael J. Fischer Lecture 16 October 30, 2017 CPSC 467, Lecture 16 1/52 Properties of Hash Functions Hash functions do not always look random Relations among

More information

Quantum Simultaneous Contract Signing

Quantum Simultaneous Contract Signing Quantum Simultaneous Contract Signing J. Bouda, M. Pivoluska, L. Caha, P. Mateus, N. Paunkovic 22. October 2010 Based on work presented on AQIS 2010 J. Bouda, M. Pivoluska, L. Caha, P. Mateus, N. Quantum

More information

Local Search (Greedy Descent): Maintain an assignment of a value to each variable. Repeat:

Local Search (Greedy Descent): Maintain an assignment of a value to each variable. Repeat: Local Search Local Search (Greedy Descent): Maintain an assignment of a value to each variable. Repeat: I I Select a variable to change Select a new value for that variable Until a satisfying assignment

More information

Distributed systems Lecture 4: Clock synchronisation; logical clocks. Dr Robert N. M. Watson

Distributed systems Lecture 4: Clock synchronisation; logical clocks. Dr Robert N. M. Watson Distributed systems Lecture 4: Clock synchronisation; logical clocks Dr Robert N. M. Watson 1 Last time Started to look at time in distributed systems Coordinating actions between processes Physical clocks

More information

Distributed Systems. Time, Clocks, and Ordering of Events

Distributed Systems. Time, Clocks, and Ordering of Events Distributed Systems Time, Clocks, and Ordering of Events Björn Franke University of Edinburgh 2016/2017 Today Last lecture: Basic Algorithms Today: Time, clocks, NTP Ref: CDK Causality, ordering, logical

More information

The Complexity of a Reliable Distributed System

The Complexity of a Reliable Distributed System The Complexity of a Reliable Distributed System Rachid Guerraoui EPFL Alexandre Maurer EPFL Abstract Studying the complexity of distributed algorithms typically boils down to evaluating how the number

More information

ACO Comprehensive Exam March 20 and 21, Computability, Complexity and Algorithms

ACO Comprehensive Exam March 20 and 21, Computability, Complexity and Algorithms 1. Computability, Complexity and Algorithms Part a: You are given a graph G = (V,E) with edge weights w(e) > 0 for e E. You are also given a minimum cost spanning tree (MST) T. For one particular edge

More information

Concurrent Non-malleable Commitments from any One-way Function

Concurrent Non-malleable Commitments from any One-way Function Concurrent Non-malleable Commitments from any One-way Function Margarita Vald Tel-Aviv University 1 / 67 Outline Non-Malleable Commitments Problem Presentation Overview DDN - First NMC Protocol Concurrent

More information

The Markov Chain Monte Carlo Method

The Markov Chain Monte Carlo Method The Markov Chain Monte Carlo Method Idea: define an ergodic Markov chain whose stationary distribution is the desired probability distribution. Let X 0, X 1, X 2,..., X n be the run of the chain. The Markov

More information

Agreement. Today. l Coordination and agreement in group communication. l Consensus

Agreement. Today. l Coordination and agreement in group communication. l Consensus Agreement Today l Coordination and agreement in group communication l Consensus Events and process states " A distributed system a collection P of N singlethreaded processes w/o shared memory Each process

More information

The Monte Carlo Method

The Monte Carlo Method The Monte Carlo Method Example: estimate the value of π. Choose X and Y independently and uniformly at random in [0, 1]. Let Pr(Z = 1) = π 4. 4E[Z] = π. { 1 if X Z = 2 + Y 2 1, 0 otherwise, Let Z 1,...,

More information

Chapter 8 The Disjoint Sets Class

Chapter 8 The Disjoint Sets Class Chapter 8 The Disjoint Sets Class 2 Introduction equivalence problem can be solved fairly simply simple data structure each function requires only a few lines of code two operations: and can be implemented

More information

Symmetric Network Computation

Symmetric Network Computation Symmetric Network Computation and the Finite-State Symmetric Graph Automaton (FSSGA) Model David Pritchard, with Santosh Vempala Theory of Distributed Systems Group at MIT, July 7 2006 David Pritchard

More information

An Algorithm for a Two-Disk Fault-Tolerant Array with (Prime 1) Disks

An Algorithm for a Two-Disk Fault-Tolerant Array with (Prime 1) Disks An Algorithm for a Two-Disk Fault-Tolerant Array with (Prime 1) Disks Sanjeeb Nanda and Narsingh Deo School of Computer Science University of Central Florida Orlando, Florida 32816-2362 sanjeeb@earthlink.net,

More information

Mixing Times and Hitting Times

Mixing Times and Hitting Times Mixing Times and Hitting Times David Aldous January 12, 2010 Old Results and Old Puzzles Levin-Peres-Wilmer give some history of the emergence of this Mixing Times topic in the early 1980s, and we nowadays

More information

arxiv: v2 [cs.ds] 3 Oct 2017

arxiv: v2 [cs.ds] 3 Oct 2017 Orthogonal Vectors Indexing Isaac Goldstein 1, Moshe Lewenstein 1, and Ely Porat 1 1 Bar-Ilan University, Ramat Gan, Israel {goldshi,moshe,porately}@cs.biu.ac.il arxiv:1710.00586v2 [cs.ds] 3 Oct 2017 Abstract

More information

Lecture 9: Counting Matchings

Lecture 9: Counting Matchings Counting and Sampling Fall 207 Lecture 9: Counting Matchings Lecturer: Shayan Oveis Gharan October 20 Disclaimer: These notes have not been subjected to the usual scrutiny reserved for formal publications.

More information

Network Algorithms and Complexity (NTUA-MPLA) Reliable Broadcast. Aris Pagourtzis, Giorgos Panagiotakos, Dimitris Sakavalas

Network Algorithms and Complexity (NTUA-MPLA) Reliable Broadcast. Aris Pagourtzis, Giorgos Panagiotakos, Dimitris Sakavalas Network Algorithms and Complexity (NTUA-MPLA) Reliable Broadcast Aris Pagourtzis, Giorgos Panagiotakos, Dimitris Sakavalas Slides are partially based on the joint work of Christos Litsas, Aris Pagourtzis,

More information

Recap. CS514: Intermediate Course in Operating Systems. What time is it? This week. Reminder: Lamport s approach. But what does time mean?

Recap. CS514: Intermediate Course in Operating Systems. What time is it? This week. Reminder: Lamport s approach. But what does time mean? CS514: Intermediate Course in Operating Systems Professor Ken Birman Vivek Vishnumurthy: TA Recap We ve started a process of isolating questions that arise in big systems Tease out an abstract issue Treat

More information

CS 188: Artificial Intelligence. Bayes Nets

CS 188: Artificial Intelligence. Bayes Nets CS 188: Artificial Intelligence Probabilistic Inference: Enumeration, Variable Elimination, Sampling Pieter Abbeel UC Berkeley Many slides over this course adapted from Dan Klein, Stuart Russell, Andrew

More information

Lecture 14 October 22

Lecture 14 October 22 EE 2: Coding for Digital Communication & Beyond Fall 203 Lecture 4 October 22 Lecturer: Prof. Anant Sahai Scribe: Jingyan Wang This lecture covers: LT Code Ideal Soliton Distribution 4. Introduction So

More information

0.1 Motivating example: weighted majority algorithm

0.1 Motivating example: weighted majority algorithm princeton univ. F 16 cos 521: Advanced Algorithm Design Lecture 8: Decision-making under total uncertainty: the multiplicative weight algorithm Lecturer: Sanjeev Arora Scribe: Sanjeev Arora (Today s notes

More information

University of Chicago Autumn 2003 CS Markov Chain Monte Carlo Methods

University of Chicago Autumn 2003 CS Markov Chain Monte Carlo Methods University of Chicago Autumn 2003 CS37101-1 Markov Chain Monte Carlo Methods Lecture 4: October 21, 2003 Bounding the mixing time via coupling Eric Vigoda 4.1 Introduction In this lecture we ll use the

More information

Our Problem. Model. Clock Synchronization. Global Predicate Detection and Event Ordering

Our Problem. Model. Clock Synchronization. Global Predicate Detection and Event Ordering Our Problem Global Predicate Detection and Event Ordering To compute predicates over the state of a distributed application Model Clock Synchronization Message passing No failures Two possible timing assumptions:

More information

Smoothed Analysis of Dynamic Networks

Smoothed Analysis of Dynamic Networks Smoothed Analysis of Dynamic Networks Michael Dinitz Johns Hopkins University mdinitz@cs.jhu.edu Seth Gilbert National University of Singapore seth.gilbert@comp.nus.edu.sg Jeremy T. Fineman Georgetown

More information

arxiv: v2 [cs.ds] 5 Apr 2014

arxiv: v2 [cs.ds] 5 Apr 2014 Simple, Fast and Deterministic Gossip and Rumor Spreading Bernhard Haeupler Microsoft Research haeupler@cs.cmu.edu arxiv:20.93v2 [cs.ds] 5 Apr 204 Abstract We study gossip algorithms for the rumor spreading

More information

1 Random Walks and Electrical Networks

1 Random Walks and Electrical Networks CME 305: Discrete Mathematics and Algorithms Random Walks and Electrical Networks Random walks are widely used tools in algorithm design and probabilistic analysis and they have numerous applications.

More information

Byzantine Agreement. Gábor Mészáros. Tatracrypt 2012, July 2 4 Smolenice, Slovakia. CEU Budapest, Hungary

Byzantine Agreement. Gábor Mészáros. Tatracrypt 2012, July 2 4 Smolenice, Slovakia. CEU Budapest, Hungary CEU Budapest, Hungary Tatracrypt 2012, July 2 4 Smolenice, Slovakia Byzantium, 1453 AD. The Final Strike... Communication via Messengers The Byzantine Generals Problem Communication System Model Goal G

More information

6.207/14.15: Networks Lecture 12: Generalized Random Graphs

6.207/14.15: Networks Lecture 12: Generalized Random Graphs 6.207/14.15: Networks Lecture 12: Generalized Random Graphs 1 Outline Small-world model Growing random networks Power-law degree distributions: Rich-Get-Richer effects Models: Uniform attachment model

More information

C 1. Recap: Finger Table. CSE 486/586 Distributed Systems Consensus. One Reason: Impossibility of Consensus. Let s Consider This

C 1. Recap: Finger Table. CSE 486/586 Distributed Systems Consensus. One Reason: Impossibility of Consensus. Let s Consider This Recap: Finger Table Finding a using fingers Distributed Systems onsensus Steve Ko omputer Sciences and Engineering University at Buffalo N102 86 + 2 4 N86 20 + 2 6 N20 2 Let s onsider This

More information

Distributed Optimization over Networks Gossip-Based Algorithms

Distributed Optimization over Networks Gossip-Based Algorithms Distributed Optimization over Networks Gossip-Based Algorithms Angelia Nedić angelia@illinois.edu ISE Department and Coordinated Science Laboratory University of Illinois at Urbana-Champaign Outline Random

More information

Spatial Gossip and Resource Location Protocols

Spatial Gossip and Resource Location Protocols Spatial Gossip and Resource Location Protocols David Kempe, Jon Kleinberg, and Alan Demers Dept. of Computer Science Cornell University, Ithaca NY 14853 email: {kempe,kleinber,ademers}@cs.cornell.edu September

More information

Lecture: Local Spectral Methods (2 of 4) 19 Computing spectral ranking with the push procedure

Lecture: Local Spectral Methods (2 of 4) 19 Computing spectral ranking with the push procedure Stat260/CS294: Spectral Graph Methods Lecture 19-04/02/2015 Lecture: Local Spectral Methods (2 of 4) Lecturer: Michael Mahoney Scribe: Michael Mahoney Warning: these notes are still very rough. They provide

More information

CSA E0 235: Cryptography (19 Mar 2015) CBC-MAC

CSA E0 235: Cryptography (19 Mar 2015) CBC-MAC CSA E0 235: Cryptography (19 Mar 2015) Instructor: Arpita Patra CBC-MAC Submitted by: Bharath Kumar, KS Tanwar 1 Overview In this lecture, we will explore Cipher Block Chaining - Message Authentication

More information

Theoretical Cryptography, Lectures 18-20

Theoretical Cryptography, Lectures 18-20 Theoretical Cryptography, Lectures 18-20 Instructor: Manuel Blum Scribes: Ryan Williams and Yinmeng Zhang March 29, 2006 1 Content of the Lectures These lectures will cover how someone can prove in zero-knowledge

More information

There are self-avoiding walks of steps on Z 3

There are self-avoiding walks of steps on Z 3 There are 7 10 26 018 276 self-avoiding walks of 38 797 311 steps on Z 3 Nathan Clisby MASCOS, The University of Melbourne Institut für Theoretische Physik Universität Leipzig November 9, 2012 1 / 37 Outline

More information

CS221 Practice Midterm #2 Solutions

CS221 Practice Midterm #2 Solutions CS221 Practice Midterm #2 Solutions Summer 2013 Updated 4:00pm, July 24 2 [Deterministic Search] Pacfamily (20 points) Pacman is trying eat all the dots, but he now has the help of his family! There are

More information

Uncertainty and Randomization

Uncertainty and Randomization Uncertainty and Randomization The PageRank Computation in Google Roberto Tempo IEIIT-CNR Politecnico di Torino tempo@polito.it 1993: Robustness of Linear Systems 1993: Robustness of Linear Systems 16 Years

More information

Trusting the Cloud with Practical Interactive Proofs

Trusting the Cloud with Practical Interactive Proofs Trusting the Cloud with Practical Interactive Proofs Graham Cormode G.Cormode@warwick.ac.uk Amit Chakrabarti (Dartmouth) Andrew McGregor (U Mass Amherst) Justin Thaler (Harvard/Yahoo!) Suresh Venkatasubramanian

More information

Randomized Algorithms

Randomized Algorithms Randomized Algorithms Prof. Tapio Elomaa tapio.elomaa@tut.fi Course Basics A new 4 credit unit course Part of Theoretical Computer Science courses at the Department of Mathematics There will be 4 hours

More information

Complexity Theory. Jörg Kreiker. Summer term Chair for Theoretical Computer Science Prof. Esparza TU München

Complexity Theory. Jörg Kreiker. Summer term Chair for Theoretical Computer Science Prof. Esparza TU München Complexity Theory Jörg Kreiker Chair for Theoretical Computer Science Prof. Esparza TU München Summer term 2010 2 Lecture 15 Public Coins and Graph (Non)Isomorphism 3 Intro Goal and Plan Goal understand

More information

1 More finite deterministic automata

1 More finite deterministic automata CS 125 Section #6 Finite automata October 18, 2016 1 More finite deterministic automata Exercise. Consider the following game with two players: Repeatedly flip a coin. On heads, player 1 gets a point.

More information

144 IEEE/ACM TRANSACTIONS ON NETWORKING, VOL. 17, NO. 1, FEBRUARY A PDF f (x) is completely monotone if derivatives f of all orders exist

144 IEEE/ACM TRANSACTIONS ON NETWORKING, VOL. 17, NO. 1, FEBRUARY A PDF f (x) is completely monotone if derivatives f of all orders exist 144 IEEE/ACM TRANSACTIONS ON NETWORKING, VOL. 17, NO. 1, FEBRUARY 2009 Node Isolation Model and Age-Based Neighbor Selection in Unstructured P2P Networks Zhongmei Yao, Student Member, IEEE, Xiaoming Wang,

More information

1 Approximate Quantiles and Summaries

1 Approximate Quantiles and Summaries CS 598CSC: Algorithms for Big Data Lecture date: Sept 25, 2014 Instructor: Chandra Chekuri Scribe: Chandra Chekuri Suppose we have a stream a 1, a 2,..., a n of objects from an ordered universe. For simplicity

More information

Homing and Synchronizing Sequences

Homing and Synchronizing Sequences Homing and Synchronizing Sequences Sven Sandberg Information Technology Department Uppsala University Sweden 1 Outline 1. Motivations 2. Definitions and Examples 3. Algorithms (a) Current State Uncertainty

More information

Evolutionary Dynamics on Graphs

Evolutionary Dynamics on Graphs Evolutionary Dynamics on Graphs Leslie Ann Goldberg, University of Oxford Absorption Time of the Moran Process (2014) with Josep Díaz, David Richerby and Maria Serna Approximating Fixation Probabilities

More information

Inaccessible Entropy and its Applications. 1 Review: Psedorandom Generators from One-Way Functions

Inaccessible Entropy and its Applications. 1 Review: Psedorandom Generators from One-Way Functions Columbia University - Crypto Reading Group Apr 27, 2011 Inaccessible Entropy and its Applications Igor Carboni Oliveira We summarize the constructions of PRGs from OWFs discussed so far and introduce the

More information

Byzantine Agreement. Gábor Mészáros. CEU Budapest, Hungary

Byzantine Agreement. Gábor Mészáros. CEU Budapest, Hungary CEU Budapest, Hungary 1453 AD, Byzantium Distibuted Systems Communication System Model Distibuted Systems Communication System Model G = (V, E) simple graph Distibuted Systems Communication System Model

More information

Hashing. Martin Babka. January 12, 2011

Hashing. Martin Babka. January 12, 2011 Hashing Martin Babka January 12, 2011 Hashing Hashing, Universal hashing, Perfect hashing Input data is uniformly distributed. A dynamic set is stored. Universal hashing Randomised algorithm uniform choice

More information

The first bound is the strongest, the other two bounds are often easier to state and compute. Proof: Applying Markov's inequality, for any >0 we have

The first bound is the strongest, the other two bounds are often easier to state and compute. Proof: Applying Markov's inequality, for any >0 we have The first bound is the strongest, the other two bounds are often easier to state and compute Proof: Applying Markov's inequality, for any >0 we have Pr (1 + ) = Pr For any >0, we can set = ln 1+ (4.4.1):

More information

Biased Quantiles. Flip Korn Graham Cormode S. Muthukrishnan

Biased Quantiles. Flip Korn Graham Cormode S. Muthukrishnan Biased Quantiles Graham Cormode cormode@bell-labs.com S. Muthukrishnan muthu@cs.rutgers.edu Flip Korn flip@research.att.com Divesh Srivastava divesh@research.att.com Quantiles Quantiles summarize data

More information

Web Structure Mining Nodes, Links and Influence

Web Structure Mining Nodes, Links and Influence Web Structure Mining Nodes, Links and Influence 1 Outline 1. Importance of nodes 1. Centrality 2. Prestige 3. Page Rank 4. Hubs and Authority 5. Metrics comparison 2. Link analysis 3. Influence model 1.

More information

A note on network reliability

A note on network reliability A note on network reliability Noga Alon Institute for Advanced Study, Princeton, NJ 08540 and Department of Mathematics Tel Aviv University, Tel Aviv, Israel Let G = (V, E) be a loopless undirected multigraph,

More information

1 Maintaining a Dictionary

1 Maintaining a Dictionary 15-451/651: Design & Analysis of Algorithms February 1, 2016 Lecture #7: Hashing last changed: January 29, 2016 Hashing is a great practical tool, with an interesting and subtle theory too. In addition

More information

Definition A finite Markov chain is a memoryless homogeneous discrete stochastic process with a finite number of states.

Definition A finite Markov chain is a memoryless homogeneous discrete stochastic process with a finite number of states. Chapter 8 Finite Markov Chains A discrete system is characterized by a set V of states and transitions between the states. V is referred to as the state space. We think of the transitions as occurring

More information

Lecture 15: MCMC Sanjeev Arora Elad Hazan. COS 402 Machine Learning and Artificial Intelligence Fall 2016

Lecture 15: MCMC Sanjeev Arora Elad Hazan. COS 402 Machine Learning and Artificial Intelligence Fall 2016 Lecture 15: MCMC Sanjeev Arora Elad Hazan COS 402 Machine Learning and Artificial Intelligence Fall 2016 Course progress Learning from examples Definition + fundamental theorem of statistical learning,

More information

CS505: Distributed Systems

CS505: Distributed Systems Cristina Nita-Rotaru CS505: Distributed Systems. Required reading for this topic } Michael J. Fischer, Nancy A. Lynch, and Michael S. Paterson for "Impossibility of Distributed with One Faulty Process,

More information

Empirical Risk Minimization

Empirical Risk Minimization Empirical Risk Minimization Fabrice Rossi SAMM Université Paris 1 Panthéon Sorbonne 2018 Outline Introduction PAC learning ERM in practice 2 General setting Data X the input space and Y the output space

More information

Lecture 1: Contraction Algorithm

Lecture 1: Contraction Algorithm CSE 5: Design and Analysis of Algorithms I Spring 06 Lecture : Contraction Algorithm Lecturer: Shayan Oveis Gharan March 8th Scribe: Mohammad Javad Hosseini Disclaimer: These notes have not been subjected

More information

1 AC 0 and Håstad Switching Lemma

1 AC 0 and Håstad Switching Lemma princeton university cos 522: computational complexity Lecture 19: Circuit complexity Lecturer: Sanjeev Arora Scribe:Tony Wirth Complexity theory s Waterloo As we saw in an earlier lecture, if PH Σ P 2

More information

Artificial Intelligence

Artificial Intelligence Artificial Intelligence Roman Barták Department of Theoretical Computer Science and Mathematical Logic Summary of last lecture We know how to do probabilistic reasoning over time transition model P(X t

More information

Graph Theorizing Peg Solitaire. D. Paul Hoilman East Tennessee State University

Graph Theorizing Peg Solitaire. D. Paul Hoilman East Tennessee State University Graph Theorizing Peg Solitaire D. Paul Hoilman East Tennessee State University December 7, 00 Contents INTRODUCTION SIMPLE SOLVING CONCEPTS 5 IMPROVED SOLVING 7 4 RELATED GAMES 5 5 PROGENATION OF SOLVABLE

More information

FINAL EXAM PRACTICE PROBLEMS CMSC 451 (Spring 2016)

FINAL EXAM PRACTICE PROBLEMS CMSC 451 (Spring 2016) FINAL EXAM PRACTICE PROBLEMS CMSC 451 (Spring 2016) The final exam will be on Thursday, May 12, from 8:00 10:00 am, at our regular class location (CSI 2117). It will be closed-book and closed-notes, except

More information

Markov Chain Monte Carlo The Metropolis-Hastings Algorithm

Markov Chain Monte Carlo The Metropolis-Hastings Algorithm Markov Chain Monte Carlo The Metropolis-Hastings Algorithm Anthony Trubiano April 11th, 2018 1 Introduction Markov Chain Monte Carlo (MCMC) methods are a class of algorithms for sampling from a probability

More information

9 Forward-backward algorithm, sum-product on factor graphs

9 Forward-backward algorithm, sum-product on factor graphs Massachusetts Institute of Technology Department of Electrical Engineering and Computer Science 6.438 Algorithms For Inference Fall 2014 9 Forward-backward algorithm, sum-product on factor graphs The previous

More information

Clock Synchronization with Bounded Global and Local Skew

Clock Synchronization with Bounded Global and Local Skew Clock Synchronization with ounded Global and Local Skew Distributed Computing Christoph Lenzen, ETH Zurich Thomas Locher, ETH Zurich Roger Wattenhofer, ETH Zurich October 2008 Motivation: No Global Clock

More information

Walk-Sum Interpretation and Analysis of Gaussian Belief Propagation

Walk-Sum Interpretation and Analysis of Gaussian Belief Propagation Walk-Sum Interpretation and Analysis of Gaussian Belief Propagation Jason K. Johnson, Dmitry M. Malioutov and Alan S. Willsky Department of Electrical Engineering and Computer Science Massachusetts Institute

More information

1 Recap: Interactive Proofs

1 Recap: Interactive Proofs Theoretical Foundations of Cryptography Lecture 16 Georgia Tech, Spring 2010 Zero-Knowledge Proofs 1 Recap: Interactive Proofs Instructor: Chris Peikert Scribe: Alessio Guerrieri Definition 1.1. An interactive

More information

CPSC 467: Cryptography and Computer Security

CPSC 467: Cryptography and Computer Security CPSC 467: Cryptography and Computer Security Michael J. Fischer Lecture 19 November 8, 2017 CPSC 467, Lecture 19 1/37 Zero Knowledge Interactive Proofs (ZKIP) ZKIP for graph isomorphism Feige-Fiat-Shamir

More information

Introduction to Algorithms March 10, 2010 Massachusetts Institute of Technology Spring 2010 Professors Piotr Indyk and David Karger Quiz 1

Introduction to Algorithms March 10, 2010 Massachusetts Institute of Technology Spring 2010 Professors Piotr Indyk and David Karger Quiz 1 Introduction to Algorithms March 10, 2010 Massachusetts Institute of Technology 6.006 Spring 2010 Professors Piotr Indyk and David Karger Quiz 1 Quiz 1 Do not open this quiz booklet until directed to do

More information

Math 4310 Solutions to homework 1 Due 9/1/16

Math 4310 Solutions to homework 1 Due 9/1/16 Math 0 Solutions to homework Due 9//6. An element [a] Z/nZ is idempotent if [a] 2 [a]. Find all idempotent elements in Z/0Z and in Z/Z. Solution. First note we clearly have [0] 2 [0] so [0] is idempotent

More information

Privacy of Numeric Queries Via Simple Value Perturbation. The Laplace Mechanism

Privacy of Numeric Queries Via Simple Value Perturbation. The Laplace Mechanism Privacy of Numeric Queries Via Simple Value Perturbation The Laplace Mechanism Differential Privacy A Basic Model Let X represent an abstract data universe and D be a multi-set of elements from X. i.e.

More information

Model Counting for Logical Theories

Model Counting for Logical Theories Model Counting for Logical Theories Wednesday Dmitry Chistikov Rayna Dimitrova Department of Computer Science University of Oxford, UK Max Planck Institute for Software Systems (MPI-SWS) Kaiserslautern

More information

Symmetric Rendezvous in Graphs: Deterministic Approaches

Symmetric Rendezvous in Graphs: Deterministic Approaches Symmetric Rendezvous in Graphs: Deterministic Approaches Shantanu Das Technion, Haifa, Israel http://www.bitvalve.org/~sdas/pres/rendezvous_lorentz.pdf Coauthors: Jérémie Chalopin, Adrian Kosowski, Peter

More information

Newscast EM. Abstract

Newscast EM. Abstract Newscast EM Wojtek Kowalczyk Department of Computer Science Vrije Universiteit Amsterdam The Netherlands wojtek@cs.vu.nl Nikos Vlassis Informatics Institute University of Amsterdam The Netherlands vlassis@science.uva.nl

More information

A Note on Google s PageRank

A Note on Google s PageRank A Note on Google s PageRank According to Google, google-search on a given topic results in a listing of most relevant web pages related to the topic. Google ranks the importance of webpages according to

More information