Rigorous Design of Moving Sequencer Atomic Broadcast with Unicast Broadcast

Size: px
Start display at page:

Download "Rigorous Design of Moving Sequencer Atomic Broadcast with Unicast Broadcast"

Transcription

1 Proc. of Int. Conf. on Advances in Computer Science, AETACS Rigorous Design of Moving Sequencer Atomic Broadcast with Unicast Broadcast Prateek Srivastava a*, Kamaljit I. Lakhtaria b and Amit Jain c a, b, c, Department of Computer Science and Engineering, Sir Padampat Singhania University,Udaipur, Rajasthan, , India prateekabhi2004@gmail.com Abstract This article investigates a novel mechanism of atomic broadcast in distributed systems. Various mechanisms are already given for moving sequencer based atomic broadcast like RMP [1], DTP [2] and Pin Wheel [3]. Since all these mechanisms are built on broadcast broadcast (BB) variant [4] of fixed sequencer atomic broadcast hence introduce a very large number of messages (2*(n-1), where n is total number of correct processes in system). Here we propose a novel mechanism that works on unicast broadcast (UB) invariant of fixed sequencer atomic broadcast in order to build moving sequencer atomic broadcast. Since the proposed mechanism relies on unicast broadcast hence it will introduce very less number of messages (n messages, where n is total number of correct processes in system) in comparison to previous mechanisms. The proposed work will reduce the load on sequencer and also provide an opportunity to each process to be a sequencer in a cyclic manner. We have used Event B [5] as formal technique for development of our model. Event B uses set theory as a modeling notation, refinements to represent system at different abstraction level. We have used Pro B [19] model checker and animator for constraint based checking, discover errors due to invariant violation and deadlocks, thereby, validating the specifications. We present an abstract model specifying atomic broadcast with a fixed sequencer and introducing moving sequencer in refined version. Index Terms Broadcast, Atomic Broadcast, Total Order, Unicast, Sequencer, Deliver, Model Checking, Event B I. INTRODUCTION Atomic broadcast (also known as total order broadcast) is an important abstraction in fault tolerant distributed computing [6]. It ensures that messages broadcasted by different processes are delivered by all destination processes in same order [7]. So, who will be responsible for ordering on messages? There are various approaches to solve this query and a lot of papers have been published in recent 25 years; The fixed sequencer atomic broadcast is one and simplest ordering mechanism. Sequencer is the site that is responsible for assigning sequence on messages before their delivery to destinations. But at high load, sequencer became overloaded and hence the performance of system degrades. Even in case of sequencer failure whole system may suffer. Hence it motivates scholars to think about alternates. Token based mechanism provides a best option as an alternate; where any process will acquire privilege before broadcasting and sequencing. Token Elsevier, 2013

2 based mechanism further characterized into two ways [4]: (i) Moving sequencer based (ii) Privilege based. For example mechanisms reported by Jia [1], Kim and Kim [2] and Flaviu et al [3] fall into first category. And mechanism reported by Richard et al [6] falls into second category. II. WHY UNICAST BROADCAST VARIANT BASED MECHANISM? There are three variants [4] of fixed sequencer total order broadcast: (i) Unicast broadcast (UB), where any process sends (unicast) its message to sequencer then sequencer put some appropriate sequence number (or order) on it and broadcast message. (ii) Broadcast broadcast (BB), where any process broadcast its message subsequently sequencer will put some sequence number (or order) on this message and broadcast. Hence there are two broadcasts for a single message. Any process can receive message only after receiving the sequence (or order) from sequencer. (iii) Unicast unicast broadcast (UUB), In this case, Any process sends (unicast) its message to sequencer then sequencer will put some appropriate sequence number (or order) on message and then return back (unicast) to sender so that the original sender can broadcast its message with a proper sequence number(or order). Any variant can be used to achieve moving sequencer total order broadcast. On scholarly observation of these moving sequencer mechanisms [1, 2, 3], it has found that all of these fall into BB category. If n is total number of correct processes then for a single message broadcast, BB generates 2*(n-1) messages; hence it leads to a very high traffic in system. Subsequently, it will degrade the performance of system. While UB based mechanism generates (n) messages that results very low traffic congestion is system and hence the performance will be maintained. It is therefore interesting to try to design atomic broadcast that rely on unicast broadcast (UB) in order to combine the advantage of UB and moving sequencer that provides, good stability time, fast delivery time and less message transmission in system. III. CONTRIBUTIONS OF THE PAPER The paper presents the first model of moving sequencer atomic broadcast that uses unicast broadcast (UB) variant of fixed sequencer atomic broadcast. The Event B; formal method is used to verify the correctness of this model. The results are obtained in sequential steps. Section 4 presents the system model for our proposed work. Section 5 concentrates on related works that already have been done by different authors; Section 6 presents the description of proposed model; this section starts with the discussion of architecture of our proposed work, then defines abstract model where fixed sequencer is used to build atomic broadcast subsequently, moving sequencer is introduced in refined model. IV. SYSTEM MODEL We assume an asynchronous system composed of n correct processes belongs to a set π = {P 1, P 2... P n }. For the simplicity we have considered a set of three processes as: Process belongs to π and Process = {P 0, P 1, P 2 }. The processes communicate by message passing over reliable channels. Message is a set of messages, for simplicity we have considered three message as: Message = {M 1, M 2, M 3 }, seq(process) is used to create a sequence on process and we have chosen [P 1, P 2, P 3 ] a sequence list for our model. Since processes are in sequence hence we denote (i) first ([P 1, P 2, P 3 ]) = {P 1 }. We also assume that there is no failure in the system. A. Agreement problem The agreement problem considered in this paper is presented below. Atomic broadcast : Atomic broadcast problem is defined by primitive [6] a_broadcast and a_delivers, the processes have to agree on a common order on a set of messages. Formally we can define atomic broadcast (uniform) by four properties [4]; (i) Validity: if a correct process a_broadcast message m then it eventually a_delivers m, (ii) Uniform agreement: If a process a_delivers m then all the correct processes a_deliver m. (iii) Uniform integrity: For any message m, every process p, a_delivers m at most once and only if m was previously a_broadcast. And (iv) Uniform total order: If some process, a_delivers m before m' then every process a_delivers m' only after it has a_delivered m. Sequencer based algorithms: The sequencer based atomic broadcast [3] is simplest one and provides best delivery time (in absence of failure) while the protocols based on privilege provide best stability time in system where logical ring is formed and message is passed along with token. In our proposed work we present a sequencer based approach, where sequencer will move in a logical ring. 485

3 V. RELATED WORK There is lot of work have been done since 25 years in area of privilege based atomic broadcast. On demand protocol [12], Train [13], Totem [14], TPM [15], Gopal and Toueg mechanism [16], RTCAST [17], MARS [18] and [6] are based on privilege. These algorithms provide a concept of logical ring in which message can travel along with token. On the other hand moving sequencer algorithms are simpler than privilege based. RMP [1], DTP [2] and Pin Wheel [3] are based on moving sequencer atomic broadcast. Our mechanism seems similar to Pin Wheel [3] but it is different in many ways. Pin Wheel [3] utilizes broadcast broadcast (BB) variant of fixed sequencer that introduces 2*(n-1) messages in system while our proposed model utilizes unicast broadcast (UB) variant of fixed sequencer that will introduce (n) messages in system which is very less in comparison to 2*(n-1). Hence, to achieve best delivery time and less message transmission in system; we have tried to present a mechanism that utilizes the advantages of moving sequencer along with unicast broadcast (UB) variant of fixed sequencer. VI. DESCRIPTION OF MODEL We have used incremental approach to design a model of total order. In first step we have developed abstract model which describes the basic functionality of total order with fixed sequencer. In refined version we have introduced moving sequencer to build total order. In most traditional moving sequencer based algorithms, broadcast broadcast (BB) variant of fixed sequencer is used [1, 2, 3]. In BB, if any process wants to broadcast their updates (messages) then it can broadcast it to all, and then sequencer process will broadcast the sequence numbers for these updates (messages). This approach (BB) introduces high message congestion into the system. A. Architecture of proposed work Fig. 1 represents the working environment of proposed work. The proposed model try to present a new approach that relies on unicast broadcast (UB) variant of fixed sequencer. In UB, if any process needs to broadcast any update (messages) then it will unicast that update to sequencer process, and then sequencer will put a unique sequence number on it and then broadcast (message with sequence number) to all. After broadcast of message, any process can deliver that message according to increasing sequence number. Before any new broadcast our model elects a new sequencer process (in refined version); which will be just next process to current sequencer. In this way it will form a logical ring. It will also reduce the load on a dedicated process that is responsible for sequencing along with its normal tasks. This approach introduces very less message (n messages) in comparison to BB (2*(n-1) messages) variant. Event B [8, 9, 10, 11] is a formal technique to develop formal models of distributed system through various refinement steps. Table 1 represents the various Event B symbols used in our model. TABLE I EVENT B SYMBOL USED IN MODEL B symbols Description : Belongs to set /: Not belongs to set <: Subset! For every X Cartesian Product POW Power Set <-> Relation PRE Pre-Condition BOOL Boolean Seq Sequence on set We have used Pro B [19] model checker and animator for B Specifications. The Pro B tool supports automatic consistency checking of B machines via model checking. It allows fully automatic animation of B specification for errors. B. Abstract model In abstract model, unicast broadcast (UB) variant of fixed sequencer is chosen to build total order. Fig. 2 summarizes all the machine variables with their corresponding initial values and constraints (or invariant) on them. 486

4 Sequencer Re selection (Next process to current sequencer will elect as new sequencer) Deliver Message (All processes will receive broadcasted messages in total order) Ordering and Broadcast (Sequencer will create an order and broadcast with message) Any new unica Unicast Message (Any process can unicast its message to sequencer) Sequencer Selection (Initially first process will be sequencer) Fig. 1. Architecture of proposed work Process and Message are two sets representing processes and messages correspondingly. selected_sequencer is variable and defined on Process set that represents sequencer process. In this model, we have assumed that, initially first process will be sequencer. sequencer_selection is a boolean flag that represents an information that any process has been elected as sequencer (if TRUE). unicast_message represents a relation between process and corresponding unicasted messages, temporary_receive represents the message delivered by current sequencer to its local memory but it is not actual delivery, follow, represents a relationship between natural numbers where (2 -> 1) represents 2 is following 1, sent is a relationship among processes, messages and sequence numbers. Sequencer will always broadcast any message with a unique sequence number. When sequencer broadcast any message it will append message with a unique sequence number. This relationship shows message and its original sender with assigned sequence number, seq_no is natural number that represents a unique sequence number provided by sequencer, receive is a relationship among processes, messages and sequence numbers. When any process receives any message then appends this information to receive set, message_with_seq_no is a relationship between message and sequencer number. This relation is maintained by sequencer to show which message is broadcasted with which sequence number. Events: Events will get enabled if all PRE conditions are true. We have designed different events that satisfy a particular purpose. Sequencer selection event: The select_sequencer event (see fig. 3) will choose a process as sequencer. Initially first process of Process set will be elected as sequencer. Unicast event: If any process wants to broadcast any message then at first it will use Unicast event (see fig. 4) to unicast its message to current sequencer. Broadcast event: Broadcast event (see fig. 5) will be used by sequencer to broadcast message with proper sequence number to all the destination processes. Hence it will broadcast (n-1) messages. Deliver event: The Deliver event (see fig. 6) will occur at every process to deliver the message. The PRE condition 487

5 MACHINE Abstract1 SETS Process= {P1, P2, P3}; Message={M1, M2, M3} VARIABLES selected_sequencer, sequencer_selection, unicast_message, temporary_receive, follow, sent, seq_no, receive, msg_with_seq_no INVARIANT selected_sequencer : POW(Process) sequencer_selection : BOOL unicast_message : Process <-> Message temporary_receive : Process <-> Message follow : NATURAL1<->NATURAL1 sent : (Process<->Message)<->NATURAL1 seq_no : NATURAL1 receive : (Process <-> Message)<-> NATURAL1 msg_with_seq_no :Message<-> NATURAL1 INITIALISATION selected_sequencer :={} sequencer_selection :=FALSE unicast_message :={} temporary_receive :={} follow :={} sent :={} seq_no :=1 receive :={} msg_with_seq_no :={} Fig. 2. Abstract Machine Sets, Variables and their initial value select_sequencer (p) = PRE p : Process & sequencer_selection = FALSE selected_sequencer := {p} sequencer_selection := TRUE Fig. 3. Sequencer Selection Event! (n, mm). (n: NATURAL1 & mm: dom (msg_with_seq_no) & (num ->n) : follow & {mm -> n} <: msg_with_seq_no=> ({p ->mm} ->n): receive) will make sure that all processes will deliver messages in total order. 488

6 Unicast (p, m) = PRE p : Process & p /: selected_sequencer & m : Message & m /: ran(unicast_message) & sequencer_selection = TRUE unicast_message := unicast_message \/ {p -> m} Fig. 4.Unicast Event Broadcast (p, m, number) = PRE p : selected_sequencer & m : ran(unicast_message) & m /: ran(temporary_receive) & number : NATURAL1 & number= seq_no temporary_receive:=temporary_receive\/{p ->m} sent := sent \/ {{p -> m} -> number} follow :=follow \/{number} X ran(sent) msg_with_seq_no :=msg_with_seq_no \/ {m -> number} seq_no :=seq_no+1 deliver(p, m, num) = Fig. 5. Broadcast Event PRE p : Process & m : dom(msg_with_seq_no) & num : ran(sent) & (m -> num) : msg_with_seq_no & ({p -> m} -> num) /: receive &! (n,mm).(n : NATURAL1 & mm : dom(msg_with_seq_no) & (num -> n) : follow & {mm -> n}<: msg_with_seq_no =>({p -> mm} - >n):receive) receive: = receive\/{{p -> m} -> num} Fig. 6. Deliver Event 489

7 C. Refined model Fig. 2 to fig. 6 presents the abstract machine where a fixed process (we have assumed first process) will work as sequencer and broadcast messages with correct and unique sequence number. In refined version (see fig. 7) of this machine we introduce the moving nature of sequencer; where next process to current sequencer will be elected as new sequencer. For this purpose we have introduced some new invariants and some new events. Variable select_new_sequencer_flag is Boolean. select_new_sequencer_flag : BOOL, If it is true, indicating that new sequencer has to be elected first. Variable sequencer_list is a sequenced list of all processes. sequencer_list : seq(process). At start of system first process will be elected as sequencer then next and so on in a cyclic manner. And process_id variable represents process s identification number that starts from 1 and increases by one. Each process has a unique identification number. process_id: NATURAL Since refined version is based on its previous version (abstract model) hence all the functionalities given in abstract model should also be available here with some new functionality that is introduced in refined version. Events: The event presented in refined version will work together with the events that are already given in abstract model. New sequencer selection event: The new event select_new_sequencer (see fig. 8) is introduced in refined to select new sequencer. It will select a process as a sequencer that is just next to current sequencer. REFINEMENT refine1 REFINES Abstract1 VARIABLES select_new_sequencer_flag, sequencer_list, process_id INVARIANT select_new_sequencer_flag : BOOL & sequencer_list : seq(process) & process_id : NATURAL INITIALISATION select_new_sequencer_flag := FALSE sequencer_list := [p1,p2,p3] process_id := 1 Fig. 7. Variables, Invariants and their initial value in refined version select_new_sequencer(p)= PRE p : Process & process_id <= size(sequencer_list) & p = sequencer_list(process_id+1) & select_new_sequencer_flag = TRUE selected_sequencer := {p} select_new_sequencer_flag := FALSE IF process_id = size(sequencer_list) - 1 process_id :=0 ELSE process_id := process_id + 1 Fig. 8. New Sequencer Selection Event 490

8 VII. CONCLUSION This paper presents a novel approach to build moving sequencer atomic broadcast that relies on unicast broadcast variant of fixed sequencer. It generates very less message (1 message for unicast + (n-1) message for broadcast = n messages) in comparison to broadcast broadcast variant (BB) of fixed sequencer. Each process will be a sequencer in a cyclic manner. So in operational phase not a single process will be overloaded continuously for sequencing. In case of message loss we can also use negative and positive acknowledgement [12] to recover it. Pro B [19], is a model checker and animator tool is used for modelling and step by step checking. We checked our model for invariant violation or for any deadlock occurrence. There is no case of invariant violation or deadlock. The model works fine with any number of processes and messages. REFERENCES [1] Jia, W., Kaiser, J., and Nett, E., RMP: Fault Tolerant GroupCommunication, Micro, IEEE, (1996); 16(2); [2] Kim, J., and Kim C., A total ordering protocol using a dynamic token-passing scheme, Distributed System Engineering, (1997), 4(2); [3] Cristian, F., Mishra, S., and Alvarez, G., High performance asynchronous atomic broadcast, Distributed System Engineering, (1997), 4(2); [4] D efago, X., Schiper, A., and Urb an, P., Total order broadcast and multicast algorithms: Taxonomy and survey, ACM Comput. Surv., (2004), 36 (4); [5] Abrial, J., R., Modeling in Event-B: System and Software Engineering, [6] Ekwall, R., and Schiper, A., A Fault-Tolerant Token-Based Atomic Broadcast Algorithm, Dependable and Secure Computing, IEEE Transactions, (2011), 8(5); [7] Hadzilacos, V., and Toueg, S., Fault-Tolerant Broadcasts and Related Problems, Distributed systems (2nd Ed.), ACM Press/ Addison- Wesley Publishing Co., (1993), [8] Yadav, D., Butler, M., Application of Event B to Global Causal Ordering for Fault Tolerant Transactions. In: REFT: Workshop on Rigorous Engineering of Fault Tolerant Systems, Newcastle upon Tyne, (2005), [9] Butler, M., Yadav, D., an Incremental Development of the Mondex System in Event-B. Formal Aspects of Computing, Springer-Verlag, (2008), 20(1); [10] Metayer, C., Abrial, J.R., Voison, L., Event-B language. Technical Report, Deliverables 3.2, EU Project IST RODIN, (2005), [11] Abrial, J.R.: A System Development Process with Event-B and the Rodin Platform. In ICFEM, (2007), Springer, LNCS, 4789; 1 3. [12] Chang, J. M., and Maxemchuk, N. F., Reliable broadcast protocols, ACM Trans. Comput. Syst., (1984), 2(3); [13] Cristian, F., Asynchronous atomic broadcast. IBM Technical Disclosure Bulletin, (1991), 33(9); [14] Amir, Y., Moser, L. E., Melliar-Smith, P. M., Agarwal, D. A., Ciarfella, P., The Totem single-ring ordering and membership protocol. ACM Trans. Comput. Syst., (1995), 13(4); [15] Rajgopalan, B., Mckinley, P., A token-based protocol for reliable, ordered multicast communication. In Proc. 8th Symp. On Reliable Distributed Systems (SRDS), (1989), [16] Gopal, A., Toueg, S., Reliable broadcast in synchronous and asynchronous Environment, In Proc. 3rd Intl. Workshop on Distributed Algorithms (WDAG). Springer-Verlag, (1989), [17] Abdelzaher, T., Shaikh, A., Jahanian, F., Shin, K., RTCAST: Lightweight multicast for real-time process groups. In IEEE Real- Time Technology and Applications Symposium, (1996), [18] Kopetz, H., Gr unsteidl, G., Reisinger, J., Fault-tolerant membership service in a synchronous distributed real-time system. In Dependable Computing and fault tolerant systems, Springer- Vienna, (1991), 4; [19] Leuschel, M., Butler, M.: Pro B: A model checker for B. In: Araki, K., Gnesi, S., Mandrioli, D. (eds.) FME, Springer,Heidelberg, (2003), LNCS, 2805;

FORMAL SPECIFICATION AND VERIFICATION OF TOTAL ORDER BROADCAST THROUGH DESTINATION AGREEMENT USING EVENT-B

FORMAL SPECIFICATION AND VERIFICATION OF TOTAL ORDER BROADCAST THROUGH DESTINATION AGREEMENT USING EVENT-B FORMAL SPECIFICATION AND VERIFICATION OF TOTAL ORDER BROADCAST THROUGH DESTINATION AGREEMENT USING EVENT-B Arun Kumar Singh 1 and Divakar Yadav 2 1 Department of Electronics Engg. IET, Lucknow,India 2

More information

A Case Study of Agreement Problems in Distributed Systems : Non-Blocking Atomic Commitment

A Case Study of Agreement Problems in Distributed Systems : Non-Blocking Atomic Commitment A Case Study of Agreement Problems in Distributed Systems : Non-Blocking Atomic Commitment Michel RAYNAL IRISA, Campus de Beaulieu 35042 Rennes Cedex (France) raynal @irisa.fr Abstract This paper considers

More information

A Mechanism for Sequential Consistency in a Distributed Objects System

A Mechanism for Sequential Consistency in a Distributed Objects System A Mechanism for Sequential Consistency in a Distributed Objects System Cristian Ţăpuş, Aleksey Nogin, Jason Hickey, and Jerome White California Institute of Technology Computer Science Department MC 256-80,

More information

High Throughput Total Order Broadcast for Cluster Environments

High Throughput Total Order Broadcast for Cluster Environments High Throughput Total Order Broadcast for Cluster Environments Rachid Guerraoui IC EPFL, Switzerland CSAIL MIT, USA Ron R. Levy IC EPFL, Switzerland Bastian Pochon IC EPFL, Switzerland Vivien Quéma INRIA,

More information

Basic vs. Reliable Multicast

Basic vs. Reliable Multicast Basic vs. Reliable Multicast Basic multicast does not consider process crashes. Reliable multicast does. So far, we considered the basic versions of ordered multicasts. What about the reliable versions?

More information

Model-based Testing Using Scenarios and Event-B Refinements

Model-based Testing Using Scenarios and Event-B Refinements Model-based Testing Using Scenarios and Event-B Refinements Qaisar A. Malik, Johan Lilius, and Linas Laibinis Åbo Akademi University, Department of Information Technologies Turku Centre for Computer Science

More information

Distributed Algorithms Reliable Broadcast

Distributed Algorithms Reliable Broadcast Distributed Algorithms Reliable Broadcast Alberto Montresor University of Trento, Italy 2016/04/26 This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. Contents

More information

Consensus Service: a modular approach for building agreement. protocols in distributed systems. Rachid Guerraoui Andre Schiper

Consensus Service: a modular approach for building agreement. protocols in distributed systems. Rachid Guerraoui Andre Schiper Consensus Service: a modular approach for building agreement protocols in distributed systems Rachid Guerraoui Andre Schiper Departement d'informatique Ecole Polytechnique Federale de Lausanne 1015 Lausanne,

More information

The Timed Asynchronous Distributed System Model By Flaviu Cristian and Christof Fetzer

The Timed Asynchronous Distributed System Model By Flaviu Cristian and Christof Fetzer The Timed Asynchronous Distributed System Model By Flaviu Cristian and Christof Fetzer - proposes a formal definition for the timed asynchronous distributed system model - presents measurements of process

More information

A Fast Group Communication Mechanism for Large Scale Distributed Objects 1

A Fast Group Communication Mechanism for Large Scale Distributed Objects 1 A Fast Group Communication Mechanism for Large Scale Distributed Objects 1 Hojjat Jafarpour and Nasser Yazdani Department of Electrical and Computer Engineering University of Tehran Tehran, Iran hjafarpour@ece.ut.ac.ir,

More information

On the interconnection of message passing systems

On the interconnection of message passing systems Information Processing Letters 105 (2008) 249 254 www.elsevier.com/locate/ipl On the interconnection of message passing systems A. Álvarez a,s.arévalo b, V. Cholvi c,, A. Fernández b,e.jiménez a a Polytechnic

More information

Distributed Algorithms Benoît Garbinato

Distributed Algorithms Benoît Garbinato Distributed Algorithms Benoît Garbinato 1 Distributed systems networks distributed As long as there were no machines, programming was no problem networks distributed at all; when we had a few weak computers,

More information

SPECIFYING AND MODELING MULTICAST COMMUNICATION IN CBCAST PROTOCOL

SPECIFYING AND MODELING MULTICAST COMMUNICATION IN CBCAST PROTOCOL THE PUBLISHING HOUSE PROCEEDINGS OF THE ROMANIAN ACADEMY, Series A, OF THE ROMANIAN ACADEMY Volume 13, Number 3/2012, pp. 261 268 SPECIFYING AND MODELING MULTICAST COMMUNICATION IN CBCAST PROTOCOL Seyed

More information

Decentralized Message Ordering

Decentralized Message Ordering Decentralized Message Ordering Cristian Lumezanu University of Maryland College Park, MD 20742 lume@cs.umd.edu advisors: Neil Spring Bobby Bhattacharjee Abstract We describe a method to order messages

More information

Site 1 Site 2 Site 3. w1[x] pos ack(c1) pos ack(c1) w2[x] neg ack(c2)

Site 1 Site 2 Site 3. w1[x] pos ack(c1) pos ack(c1) w2[x] neg ack(c2) Using Broadcast Primitives in Replicated Databases y I. Stanoi D. Agrawal A. El Abbadi Dept. of Computer Science University of California Santa Barbara, CA 93106 E-mail: fioana,agrawal,amrg@cs.ucsb.edu

More information

Active leave behavior of members in a fault-tolerant group

Active leave behavior of members in a fault-tolerant group 260 Science in China Ser. F Information Sciences 2004 Vol.47 No.2 260 272 Active leave behavior of members in a fault-tolerant group WANG Yun Department of Computer Science and Engineering, Southeast University,

More information

An Analysis and Improvement of Probe-Based Algorithm for Distributed Deadlock Detection

An Analysis and Improvement of Probe-Based Algorithm for Distributed Deadlock Detection An Analysis and Improvement of Probe-Based Algorithm for Distributed Deadlock Detection Kunal Chakma, Anupam Jamatia, and Tribid Debbarma Abstract In this paper we have performed an analysis of existing

More information

Finding Deadlocks of Event-B Models by Constraint Solving

Finding Deadlocks of Event-B Models by Constraint Solving B 2011 June 21st, 2011, Limerick, Ireland Finding Deadlocks of Event-B Models by Constraint Solving Stefan Hallerstede, Michael Leuschel Institut für Informatik, Universität Düsseldorf Universitätsstr.

More information

A Group Communication Protocol for CORBA

A Group Communication Protocol for CORBA A Group Communication Protocol for CORBA L. E. Moser, P. M. Melliar-Smith, R. Koch, K. Berket Department of Electrical and Computer Engineering University of California, Santa Barbara 93106 Abstract Group

More information

Ruminations on Domain-Based Reliable Broadcast

Ruminations on Domain-Based Reliable Broadcast Ruminations on Domain-Based Reliable Broadcast Svend Frølund Fernando Pedone Hewlett-Packard Laboratories Palo Alto, CA 94304, USA Abstract A distributed system is no longer confined to a single administrative

More information

Verification of UML models by translation to UML-B

Verification of UML models by translation to UML-B Verification of UML models by translation to UML-B Colin Snook, Vitaly Savicks and Michael Butler University of Southampton Abstract. UML-B is a UML like notation based on the Event-B formalism which allows

More information

Coordination and Agreement

Coordination and Agreement Coordination and Agreement 12.1 Introduction 12.2 Distributed Mutual Exclusion 12.4 Multicast Communication 12.3 Elections 12.5 Consensus and Related Problems AIM: Coordination and/or Agreement Collection

More information

Formal Development of a Total Order Broadcast for Distributed Transactions using Event-B

Formal Development of a Total Order Broadcast for Distributed Transactions using Event-B Formal Development of a Total Order Broadcast for Distributed Transactions using Event-B Divakar Yadav and Michael Butler School of Electronics and Computer Science University of Southampton Southampton

More information

Selected Questions. Exam 2 Fall 2006

Selected Questions. Exam 2 Fall 2006 Selected Questions Exam 2 Fall 2006 Page 1 Question 5 The clock in the clock tower in the town of Chronos broke. It was repaired but now the clock needs to be set. A train leaves for the nearest town,

More information

Coordination and Agreement

Coordination and Agreement Coordination and Agreement Nicola Dragoni Embedded Systems Engineering DTU Informatics 1. Introduction 2. Distributed Mutual Exclusion 3. Elections 4. Multicast Communication 5. Consensus and related problems

More information

Coordination 2. Today. How can processes agree on an action or a value? l Group communication l Basic, reliable and l ordered multicast

Coordination 2. Today. How can processes agree on an action or a value? l Group communication l Basic, reliable and l ordered multicast Coordination 2 Today l Group communication l Basic, reliable and l ordered multicast How can processes agree on an action or a value? Modes of communication Unicast 1ç è 1 Point to point Anycast 1è

More information

Specifying and Proving Broadcast Properties with TLA

Specifying and Proving Broadcast Properties with TLA Specifying and Proving Broadcast Properties with TLA William Hipschman Department of Computer Science The University of North Carolina at Chapel Hill Abstract Although group communication is vitally important

More information

Process groups and message ordering

Process groups and message ordering Process groups and message ordering If processes belong to groups, certain algorithms can be used that depend on group properties membership create ( name ), kill ( name ) join ( name, process ), leave

More information

An optimal novel Byzantine agreement protocol (ONBAP) for heterogeneous distributed database processing systems

An optimal novel Byzantine agreement protocol (ONBAP) for heterogeneous distributed database processing systems Available online at www.sciencedirect.com Procedia Technology 6 (2012 ) 57 66 2 nd International Conference on Communication, Computing & Security An optimal novel Byzantine agreement protocol (ONBAP)

More information

Hebrew University. Jerusalem. Israel. Abstract. Transis is a high availability distributed system, being developed

Hebrew University. Jerusalem. Israel. Abstract. Transis is a high availability distributed system, being developed The Design of the Transis System??? Danny Dolev??? and Dalia Malki y Computer Science Institute Hebrew University Jerusalem Israel Abstract. Transis is a high availability distributed system, being developed

More information

Accepted Manuscript. On the Interconnection of Message Passing Systems. A. Álvarez, S. Arévalo, V. Cholvi, A. Fernández, E.

Accepted Manuscript. On the Interconnection of Message Passing Systems. A. Álvarez, S. Arévalo, V. Cholvi, A. Fernández, E. Accepted Manuscript On the Interconnection of Message Passing Systems A. Álvarez, S. Arévalo, V. Cholvi, A. Fernández, E. Jiménez PII: S0020-0190(07)00270-0 DOI: 10.1016/j.ipl.2007.09.006 Reference: IPL

More information

Throughput Optimal Total Order Broadcast for Cluster Environments 1

Throughput Optimal Total Order Broadcast for Cluster Environments 1 Throughput Optimal Total Order Broadcast for Cluster Environments 1 RACHID GUERRAOUI, RON R. LEVY, BASTIAN POCHON School of Computer and Communication Sciences, EPFL and VIVIEN QUÉMA Centre National de

More information

Composing Distributed Fault-tolerance Components

Composing Distributed Fault-tolerance Components Composing Distributed Fault-tolerance Components Sandeep S. Kulkarni Karun N. Biyani Umamaheswaran Arumugam Department of Computer Science and Engineering Michigan State University East Lansing, MI 48824

More information

AN EFFICIENT DESIGN OF VLSI ARCHITECTURE FOR FAULT DETECTION USING ORTHOGONAL LATIN SQUARES (OLS) CODES

AN EFFICIENT DESIGN OF VLSI ARCHITECTURE FOR FAULT DETECTION USING ORTHOGONAL LATIN SQUARES (OLS) CODES AN EFFICIENT DESIGN OF VLSI ARCHITECTURE FOR FAULT DETECTION USING ORTHOGONAL LATIN SQUARES (OLS) CODES S. SRINIVAS KUMAR *, R.BASAVARAJU ** * PG Scholar, Electronics and Communication Engineering, CRIT

More information

Formal Development of Fault Tolerant Transactions for a Replicated Database using Ordered Broadcasts

Formal Development of Fault Tolerant Transactions for a Replicated Database using Ordered Broadcasts Formal Development of Fault Tolerant Transactions for a Replicated Database using Ordered Broadcasts Divakar Yadav and Michael Butler Dependable Systems and Software Engineering School of Electronics and

More information

Coordination 1. To do. Mutual exclusion Election algorithms Next time: Global state. q q q

Coordination 1. To do. Mutual exclusion Election algorithms Next time: Global state. q q q Coordination 1 To do q q q Mutual exclusion Election algorithms Next time: Global state Coordination and agreement in US Congress 1798-2015 Process coordination How can processes coordinate their action?

More information

Implementation of Clocks and Sensors

Implementation of Clocks and Sensors Implementation of Clocks and Sensors Term Paper EE 382N Distributed Systems Dr. Garg November 30, 2000 Submitted by: Yousuf Ahmed Chandresh Jain Onur Mutlu Global Predicate Detection in Distributed Systems

More information

DISTRIBUTED HASH TABLE PROTOCOL DETECTION IN WIRELESS SENSOR NETWORKS

DISTRIBUTED HASH TABLE PROTOCOL DETECTION IN WIRELESS SENSOR NETWORKS DISTRIBUTED HASH TABLE PROTOCOL DETECTION IN WIRELESS SENSOR NETWORKS Mr. M. Raghu (Asst.professor) Dr.Pauls Engineering College Ms. M. Ananthi (PG Scholar) Dr. Pauls Engineering College Abstract- Wireless

More information

Replicated Database Recovery using Multicast Communication

Replicated Database Recovery using Multicast Communication Replicated Database Recovery using Multicast Communication JoAnne Holliday Dept of Computer Engineering Santa Clara University, Santa Clara, CA 95053 jholliday@acm.org Abstract Database replication with

More information

Novel low-overhead roll-forward recovery scheme for distributed systems

Novel low-overhead roll-forward recovery scheme for distributed systems Novel low-overhead roll-forward recovery scheme for distributed systems B. Gupta, S. Rahimi and Z. Liu Abstract: An efficient roll-forward checkpointing/recovery scheme for distributed systems has been

More information

Evaluating Certification Protocols in the Partial Database State Machine

Evaluating Certification Protocols in the Partial Database State Machine Evaluating Certification Protocols in the Partial Database State Machine A. Sousa, A. Correia Jr., F. Moura, J. Pereira, R. Oliveira Abstract Partial replication is an alluring technique to ensure the

More information

Synchronization. Chapter 5

Synchronization. Chapter 5 Synchronization Chapter 5 Clock Synchronization In a centralized system time is unambiguous. (each computer has its own clock) In a distributed system achieving agreement on time is not trivial. (it is

More information

Proving the Correctness of Distributed Algorithms using TLA

Proving the Correctness of Distributed Algorithms using TLA Proving the Correctness of Distributed Algorithms using TLA Khushboo Kanjani, khush@cs.tamu.edu, Texas A & M University 11 May 2007 Abstract This work is a summary of the Temporal Logic of Actions(TLA)

More information

6. The Bounded Re-transmission Protocol. Jean-Raymond Abrial

6. The Bounded Re-transmission Protocol. Jean-Raymond Abrial 6. The Bounded Re-transmission Protocol Jean-Raymond Abrial 2009 Purpose of this Lecture 1 - The Bounded Re-transmission Protocol is a file transfer protocol - This is a problem dealing with fault tolerance

More information

Packing Messages as a Tool for Boosting the Performance of. Roy Friedman Robbert van Renesse. Cornell University. Abstract

Packing Messages as a Tool for Boosting the Performance of. Roy Friedman Robbert van Renesse. Cornell University. Abstract Packing Messages as a Tool for Boosting the Performance of Total Ordering Protocols Roy Friedman Robbert van Renesse Department of Computer Science Cornell University Ithaca, NY 14853. July 7, 1995 Abstract

More information

Distributed Systems. 09. State Machine Replication & Virtual Synchrony. Paul Krzyzanowski. Rutgers University. Fall Paul Krzyzanowski

Distributed Systems. 09. State Machine Replication & Virtual Synchrony. Paul Krzyzanowski. Rutgers University. Fall Paul Krzyzanowski Distributed Systems 09. State Machine Replication & Virtual Synchrony Paul Krzyzanowski Rutgers University Fall 2016 1 State machine replication 2 State machine replication We want high scalability and

More information

Lecture 12: Time Distributed Systems

Lecture 12: Time Distributed Systems Lecture 12: Time Distributed Systems Behzad Bordbar School of Computer Science, University of Birmingham, UK Lecture 12 1 Overview Time service requirements and problems sources of time Clock synchronisation

More information

A Comparative Study of Lossless Compression Algorithm on Text Data

A Comparative Study of Lossless Compression Algorithm on Text Data Proc. of Int. Conf. on Advances in Computer Science, AETACS A Comparative Study of Lossless Compression Algorithm on Text Data Amit Jain a * Kamaljit I. Lakhtaria b, Prateek Srivastava c a, b, c Department

More information

Efficient Recovery in Harp

Efficient Recovery in Harp Efficient Recovery in Harp Barbara Liskov Sanjay Ghemawat Robert Gruber Paul Johnson Liuba Shrira Laboratory for Computer Science Massachusetts Institute of Technology 1. Introduction Harp is a replicated

More information

RTCAST: Lightweight Multicast for Real-Time Process Groups

RTCAST: Lightweight Multicast for Real-Time Process Groups RTCAST: Lightweight Multicast for Real-Time Process Groups Tarek Abdelzaher, Anees Shaikh, Farnam Jahanian, and Kang Shin Real-time Computing Laboratory Department of Electrical Engineering and Computer

More information

Using a Managed Peer-to-Peer Protocol to Deliver Files within an Organization

Using a Managed Peer-to-Peer Protocol to Deliver Files within an Organization ASIA-PACIFIC SYMPOSIUM ON QUEUEING THEORY AND NETWORK APPLICATIONS 2007 227 Using a Managed Peer-to-Peer Protocol to Deliver s within an Organization Chih-Chin Liang a, Chia-Hung Wang b, Hsing Luh b and

More information

The Totem System. L. E. Moser, P. M. Melliar-Smith, D. A. Agarwal, R. K. Budhia, C. A. Lingley-Papadopoulos, T. P. Archambault

The Totem System. L. E. Moser, P. M. Melliar-Smith, D. A. Agarwal, R. K. Budhia, C. A. Lingley-Papadopoulos, T. P. Archambault The Totem System L. E. Moser, P. M. Melliar-Smith, D. A. Agarwal, R. K. Budhia, C. A. Lingley-Papadopoulos, T. P. Archambault Department of Electrical and Computer Engineering University of California,

More information

BARP-A Dynamic Routing Protocol for Balanced Distribution of Traffic in NoCs

BARP-A Dynamic Routing Protocol for Balanced Distribution of Traffic in NoCs -A Dynamic Routing Protocol for Balanced Distribution of Traffic in NoCs Pejman Lotfi-Kamran, Masoud Daneshtalab *, Caro Lucas, and Zainalabedin Navabi School of Electrical and Computer Engineering, The

More information

UMCS. Annales UMCS Informatica AI 6 (2007) Fault tolerant control for RP* architecture of Scalable Distributed Data Structures

UMCS. Annales UMCS Informatica AI 6 (2007) Fault tolerant control for RP* architecture of Scalable Distributed Data Structures Annales Informatica AI 6 (2007) 5-13 Annales Informatica Lublin-Polonia Sectio AI http://www.annales.umcs.lublin.pl/ Fault tolerant control for RP* architecture of Scalable Distributed Data Structures

More information

Computer Science Technical Report

Computer Science Technical Report Computer Science Technical Report Feasibility of Stepwise Addition of Multitolerance to High Atomicity Programs Ali Ebnenasir and Sandeep S. Kulkarni Michigan Technological University Computer Science

More information

Lecture 13: Interconnection Networks. Topics: lots of background, recent innovations for power and performance

Lecture 13: Interconnection Networks. Topics: lots of background, recent innovations for power and performance Lecture 13: Interconnection Networks Topics: lots of background, recent innovations for power and performance 1 Interconnection Networks Recall: fully connected network, arrays/rings, meshes/tori, trees,

More information

Lecture 10: Clocks and Time

Lecture 10: Clocks and Time 06-06798 Distributed Systems Lecture 10: Clocks and Time Distributed Systems 1 Time service Overview requirements and problems sources of time Clock synchronisation algorithms clock skew & drift Cristian

More information

Implementing Synchronous Counter using Data Mining Techniques

Implementing Synchronous Counter using Data Mining Techniques Implementing Synchronous Counter using Data Mining Techniques Sangeetha S Assistant Professor,Department of Computer Science and Engineering, B.N.M Institute of Technology, Bangalore, Karnataka, India

More information

Formal Verification of Distributed Transaction Execution in Replicated Database System using Event - B

Formal Verification of Distributed Transaction Execution in Replicated Database System using Event - B Formal Verification of Distributed Transaction Execution in Replicated Database System using Event - B Pooja Sharma Pranveer Singh Institute of Technology, Kanpur, U.P. (208020) U.P.T.U., Lucknow Raghuraj

More information

Backup segments. Path after failure recovery. Fault. Primary channel. Initial path D1 D2. Primary channel 1. Backup channel 1.

Backup segments. Path after failure recovery. Fault. Primary channel. Initial path D1 D2. Primary channel 1. Backup channel 1. A Segmented Backup Scheme for Dependable Real Time Communication in Multihop Networks Gummadi P. Krishna M. Jnana Pradeep and C. Siva Ram Murthy Department of Computer Science and Engineering Indian Institute

More information

A Comparative study of On-Demand Data Delivery with Tables Driven and On-Demand Protocols for Mobile Ad-Hoc Network

A Comparative study of On-Demand Data Delivery with Tables Driven and On-Demand Protocols for Mobile Ad-Hoc Network A Comparative study of On-Demand Data Delivery with Tables Driven and On-Demand Protocols for Mobile Ad-Hoc Network Humayun Bakht Research Fellow, London School of Commerce, United Kingdom humayunbakht@yahoo.co.uk

More information

Dynamic Broadcast Scheduling in DDBMS

Dynamic Broadcast Scheduling in DDBMS Dynamic Broadcast Scheduling in DDBMS Babu Santhalingam #1, C.Gunasekar #2, K.Jayakumar #3 #1 Asst. Professor, Computer Science and Applications Department, SCSVMV University, Kanchipuram, India, #2 Research

More information

Distributed Systems. coordination Johan Montelius ID2201. Distributed Systems ID2201

Distributed Systems. coordination Johan Montelius ID2201. Distributed Systems ID2201 Distributed Systems ID2201 coordination Johan Montelius 1 Coordination Coordinating several threads in one node is a problem, coordination in a network is of course worse: failure of nodes and networks

More information

Distributed Systems (5DV147)

Distributed Systems (5DV147) Distributed Systems (5DV147) Mutual Exclusion and Elections Fall 2013 1 Processes often need to coordinate their actions Which rocess gets to access a shared resource? Has the master crashed? Elect a new

More information

2.1 CHANNEL ALLOCATION 2.2 MULTIPLE ACCESS PROTOCOLS Collision Free Protocols 2.3 FDDI 2.4 DATA LINK LAYER DESIGN ISSUES 2.5 FRAMING & STUFFING

2.1 CHANNEL ALLOCATION 2.2 MULTIPLE ACCESS PROTOCOLS Collision Free Protocols 2.3 FDDI 2.4 DATA LINK LAYER DESIGN ISSUES 2.5 FRAMING & STUFFING UNIT-2 2.1 CHANNEL ALLOCATION 2.2 MULTIPLE ACCESS PROTOCOLS 2.2.1 Pure ALOHA 2.2.2 Slotted ALOHA 2.2.3 Carrier Sense Multiple Access 2.2.4 CSMA with Collision Detection 2.2.5 Collision Free Protocols 2.2.5.1

More information

A Two-Layer Hybrid Algorithm for Achieving Mutual Exclusion in Distributed Systems

A Two-Layer Hybrid Algorithm for Achieving Mutual Exclusion in Distributed Systems A Two-Layer Hybrid Algorithm for Achieving Mutual Exclusion in Distributed Systems QUAZI EHSANUL KABIR MAMUN *, MORTUZA ALI *, SALAHUDDIN MOHAMMAD MASUM, MOHAMMAD ABDUR RAHIM MUSTAFA * Dept. of CSE, Bangladesh

More information

DESIGN AND IMPLEMENTATION OF A CORBA FAULT-TOLERANT OBJECT GROUP SERVICE

DESIGN AND IMPLEMENTATION OF A CORBA FAULT-TOLERANT OBJECT GROUP SERVICE DESIGN AND IMPLEMENTATION OF A CORBA FAULT-TOLERANT OBJECT GROUP SERVICE G. Morgan, S.K. Shrivastava, P.D. Ezhilchelvan and M.C. Little ABSTRACT Department of Computing Science, Newcastle University, Newcastle

More information

Distributed systems. Lecture 6: distributed transactions, elections, consensus and replication. Malte Schwarzkopf

Distributed systems. Lecture 6: distributed transactions, elections, consensus and replication. Malte Schwarzkopf Distributed systems Lecture 6: distributed transactions, elections, consensus and replication Malte Schwarzkopf Last time Saw how we can build ordered multicast Messages between processes in a group Need

More information

SIMULATING I/O AUTOMATA. TRACE PROPERTIES

SIMULATING I/O AUTOMATA. TRACE PROPERTIES SIMULATING I/O AUTOMATA. TRACE PROPERTIES Cristian Gavrilă and Ioan Jurca fcristig,ionelg@cs.utt.ro University Politehnica of Timişoara Computer Science Department Abstract I/O automaton is a mathematical

More information

Equations for Asynchronous Message Passing

Equations for Asynchronous Message Passing Equations for Asynchronous Message Passing Ludwik Czaja 1,2 1 Institute of Informatics, The University of Warsaw 2 University of Economics and Computer Science Vistula in Warsaw lczaja@mimuw.edu.pl 1 Introduction

More information

A Robust Bloom Filter

A Robust Bloom Filter A Robust Bloom Filter Yoon-Hwa Choi Department of Computer Engineering, Hongik University, Seoul, Korea. Orcid: 0000-0003-4585-2875 Abstract A Bloom filter is a space-efficient randomized data structure

More information

6. Node Disjoint Split Multipath Protocol for Unified. Multicasting through Announcements (NDSM-PUMA)

6. Node Disjoint Split Multipath Protocol for Unified. Multicasting through Announcements (NDSM-PUMA) 103 6. Node Disjoint Split Multipath Protocol for Unified Multicasting through Announcements (NDSM-PUMA) 6.1 Introduction It has been demonstrated in chapter 3 that the performance evaluation of the PUMA

More information

Reliable Unicasting in Faulty Hypercubes Using Safety Levels

Reliable Unicasting in Faulty Hypercubes Using Safety Levels IEEE TRANSACTIONS ON COMPUTERS, VOL. 46, NO. 2, FEBRUARY 997 24 Reliable Unicasting in Faulty Hypercubes Using Safety Levels Jie Wu, Senior Member, IEEE Abstract We propose a unicasting algorithm for faulty

More information

Distributed Systems Multicast & Group Communication Services

Distributed Systems Multicast & Group Communication Services Distributed Systems 600.437 Multicast & Group Communication Services Department of Computer Science The Johns Hopkins University 1 Multicast & Group Communication Services Lecture 3 Guide to Reliable Distributed

More information

Consensus in Byzantine Asynchronous Systems

Consensus in Byzantine Asynchronous Systems Consensus in Byzantine Asynchronous Systems R. BALDONI Universitá La Sapienza, Roma, Italy J.-M. HÉLARY IRISA, Université de Rennes 1, France M. RAYNAL IRISA, Université de Rennes 1, France L. TANGUY IRISA,

More information

Distributed Systems. replication Johan Montelius ID2201. Distributed Systems ID2201

Distributed Systems. replication Johan Montelius ID2201. Distributed Systems ID2201 Distributed Systems ID2201 replication Johan Montelius 1 The problem The problem we have: servers might be unavailable The solution: keep duplicates at different servers 2 Building a fault-tolerant service

More information

Delivery of Consistent and Integrated User s Data within a Multi-Tenant Adaptive SaaS Application

Delivery of Consistent and Integrated User s Data within a Multi-Tenant Adaptive SaaS Application Delivery of Consistent and Integrated User s Data within a Multi-Tenant Adaptive SaaS Application Aivars Niedritis University of Latvia, Faculty of Computer Science, Raina boulv.19, Riga Aivars.Niedritis@lu.lv

More information

Method and case study of model checking concurrent systems that use unbounded timestamps

Method and case study of model checking concurrent systems that use unbounded timestamps 2017 IEEE 22nd Pacific Rim International Symposium on Dependable Computing Method and case study of model checking concurrent systems that use unbounded timestamps Shinya Nakano Graduate School of Information

More information

Causal Order Multicast Protocol Using Different Information from Brokers to Subscribers

Causal Order Multicast Protocol Using Different Information from Brokers to Subscribers , pp.15-19 http://dx.doi.org/10.14257/astl.2014.51.04 Causal Order Multicast Protocol Using Different Information from Brokers to Subscribers Chayoung Kim 1 and Jinho Ahn 1, 1 Dept. of Comp. Scie., Kyonggi

More information

Hunting for Bindings in Distributed Object-Oriented Systems

Hunting for Bindings in Distributed Object-Oriented Systems Hunting for Bindings in Distributed Object-Oriented Systems Magdalena S lawiñska Faculty of Electronics, Telecommunications and Informatics Gdańsk University of Technology Narutowicza 11/12, 80-952 Gdańsk,

More information

A CAN-Based Architecture for Highly Reliable Communication Systems

A CAN-Based Architecture for Highly Reliable Communication Systems A CAN-Based Architecture for Highly Reliable Communication Systems H. Hilmer Prof. Dr.-Ing. H.-D. Kochs Gerhard-Mercator-Universität Duisburg, Germany E. Dittmar ABB Network Control and Protection, Ladenburg,

More information

Parallel Programming

Parallel Programming Parallel Programming for Multicore and Cluster Systems von Thomas Rauber, Gudula Rünger 1. Auflage Parallel Programming Rauber / Rünger schnell und portofrei erhältlich bei beck-shop.de DIE FACHBUCHHANDLUNG

More information

Applications of Paxos Algorithm

Applications of Paxos Algorithm Applications of Paxos Algorithm Gurkan Solmaz COP 6938 - Cloud Computing - Fall 2012 Department of Electrical Engineering and Computer Science University of Central Florida - Orlando, FL Oct 15, 2012 1

More information

Design and Implementation of a Consistent Time Service for Fault-Tolerant Distributed Systems

Design and Implementation of a Consistent Time Service for Fault-Tolerant Distributed Systems Design and Implementation of a Consistent Time Service for Fault-Tolerant Distributed Systems W. Zhao, L. E. Moser and P. M. Melliar-Smith Eternal Systems, Inc. 5290 Overpass Road, Building D, Santa Barbara,

More information

An Efficient Silent Self-Stabilizing Algorithm for 1-Maximal Matching in Anonymous Network

An Efficient Silent Self-Stabilizing Algorithm for 1-Maximal Matching in Anonymous Network An Efficient Silent Self-Stabilizing Algorithm for 1-Maximal Matching in Anonymous Network Yuma Asada and Michiko Inoue Nara Institute of Science and Technology, 8916-5 Takayama, Ikoma, NARA 630-0192 JAPAN,

More information

Engineering shortest-path algorithms for dynamic networks

Engineering shortest-path algorithms for dynamic networks Engineering shortest-path algorithms for dynamic networks Mattia D Emidio and Daniele Frigioni Department of Information Engineering, Computer Science and Mathematics, University of L Aquila, Via Gronchi

More information

Multimedia Multicast Transport Service for Groupware

Multimedia Multicast Transport Service for Groupware Multimedia Multicast Transport Service for Groupware Chockler, Gregory V., Huleihel, Nabil, Keidar, Idit, and Dolev, Danny, The Hebrew University of Jerusalem, Jerusalem, Israel 1.0 Abstract Reliability

More information

Comparison of Failure Detectors and Group Membership: Performance Study of Two Atomic Broadcast Algorithms

Comparison of Failure Detectors and Group Membership: Performance Study of Two Atomic Broadcast Algorithms Comparison of Failure Detectors and Group Membership: Performance Study of Two Atomic Broadcast Algorithms Péter Urbán peter.urban@epfl.ch Ilya Shnayderman ilia@cs.huji.ac.il André Schiper andre.schiper@epfl.ch

More information

Information Exchange in Wireless Networks with Network Coding and Physical-layer Broadcast

Information Exchange in Wireless Networks with Network Coding and Physical-layer Broadcast 2005 Conference on Information Sciences and Systems, The Johns Hopkins University, March 16 18, 2005 Information Exchange in Wireless Networks with Network Coding and Physical-layer Broadcast Yunnan Wu

More information

An Iterative Greedy Approach Using Geographical Destination Routing In WSN

An Iterative Greedy Approach Using Geographical Destination Routing In WSN ISSN (Online) : 2319-8753 ISSN (Print) : 2347-6710 International Journal of Innovative Research in Science, Engineering and Technology Volume 3, Special Issue 3, March 2014 2014 International Conference

More information

Introduction to Distributed Systems Seif Haridi

Introduction to Distributed Systems Seif Haridi Introduction to Distributed Systems Seif Haridi haridi@kth.se What is a distributed system? A set of nodes, connected by a network, which appear to its users as a single coherent system p1 p2. pn send

More information

6 Distributed data management I Hashing

6 Distributed data management I Hashing 6 Distributed data management I Hashing There are two major approaches for the management of data in distributed systems: hashing and caching. The hashing approach tries to minimize the use of communication

More information

Message Passing Models and Multicomputer distributed system LECTURE 7

Message Passing Models and Multicomputer distributed system LECTURE 7 Message Passing Models and Multicomputer distributed system LECTURE 7 DR SAMMAN H AMEEN 1 Node Node Node Node Node Node Message-passing direct network interconnection Node Node Node Node Node Node PAGE

More information

International Journal of Advanced Research in Computer Science and Software Engineering

International Journal of Advanced Research in Computer Science and Software Engineering Volume 2, Issue 9, September 2012 ISSN: 2277 128X International Journal of Advanced Research in Computer Science and Software Engineering Research Paper Available online at: www.ijarcsse.com Backup Two

More information

Cache Coherence in Distributed and Replicated Transactional Memory Systems. Technical Report RT/4/2009

Cache Coherence in Distributed and Replicated Transactional Memory Systems. Technical Report RT/4/2009 Technical Report RT/4/2009 Cache Coherence in Distributed and Replicated Transactional Memory Systems Maria Couceiro INESC-ID/IST maria.couceiro@ist.utl.pt Luis Rodrigues INESC-ID/IST ler@ist.utl.pt Jan

More information

A Membership Protocol for Multi-Computer Clusters

A Membership Protocol for Multi-Computer Clusters A Membership Protocol for Multi-Computer Clusters Francesc D. Muñoz-Escoí Vlada Matena José M. Bernabéu-Aubán Pablo Galdámez Technical Report ITI-ITE-98/4 Abstract Distributed applications need membership

More information

IMS signalling for multiparty services based on network level multicast

IMS signalling for multiparty services based on network level multicast IMS signalling for multiparty services based on network level multicast Ivan Vidal, Ignacio Soto, Francisco Valera, Jaime Garcia, Arturo Azcorra UniversityCarlosIIIofMadrid Av.Universidad,30 E-28911, Madrid,

More information

Contents. References 43

Contents. References 43 Contents 1 Atomicity Decomposition Part 1 - Overview and Background 1 1.1 Introduction................................... 1 1.2 Overview of Atomicity Decomposition Diagram in Event-B........ 1 1.3 Event-B

More information

殷亚凤. Consistency and Replication. Distributed Systems [7]

殷亚凤. Consistency and Replication. Distributed Systems [7] Consistency and Replication Distributed Systems [7] 殷亚凤 Email: yafeng@nju.edu.cn Homepage: http://cs.nju.edu.cn/yafeng/ Room 301, Building of Computer Science and Technology Review Clock synchronization

More information

Semi-Passive Replication in the Presence of Byzantine Faults

Semi-Passive Replication in the Presence of Byzantine Faults Semi-Passive Replication in the Presence of Byzantine Faults HariGovind V. Ramasamy Adnan Agbaria William H. Sanders University of Illinois at Urbana-Champaign 1308 W. Main Street, Urbana IL 61801, USA

More information