Skip to main content

Luis Rodrigues

Followers

0

Following

1

Public Views

Kansas State University

Laura E . Taylor

York University

Zlata Vuksanović-Macura

Geographical Institute "Jovan Cvijic" of Serbian Academy of Arts and Sciences

Johns Hopkins University

Heidi Jane Smith

Universidad Iberoamericana - Mexico

Università degli Studi "La Sapienza" di Roma

Mimar Sinan Güzel Sanatlar

Homero R Saltalamacchia

Universidad Nacional de Tres de Febrero

Konstantin Grigorichev

Irkutsk State University

Willow Lung-Amam

University of Maryland

Interests

Uploads

Papers by Luis Rodrigues

On Replication of Software Transactional Memories

Lecture Notes in Computer Science, 2008

Software Transactional Memory (STM) systems have garnered considerable interest of late due to th... more Software Transactional Memory (STM) systems have garnered considerable interest of late due to the recent architectural trend that has led to the pervasive adoption of multi-core CPUs. STMs represent an attractive solution to spare programmers from the pitfalls of conventional explicit lock-based thread synchronization, leveraging on concurrency-control concepts used for decades by the database community to simplify the mainstream parallel programming [1].

Fast Localized Delaunay Triangulation

Lecture Notes in Computer Science, 2005

A localized Delaunay triangulation owns the following interesting properties in a wireless ad hoc... more A localized Delaunay triangulation owns the following interesting properties in a wireless ad hoc setting: it can be built with localized information, the communication cost imposed by control information is limited and it supports geographical routing algorithms that offer guaranteed convergence. This paper presents a localized algorithm that builds a graph called planar localized Delaunay triangulation, P LDel, known to be a good spanner of the unit disk graph, U DG. Unlike previous work, our algorithm builds P LDel in a single communication step, maintaining a communication cost of O(n log n), which is within a constant of the optimum. This represents a significant practical improvement over previous algorithms with similar theoretical bounds. Furthermore, the small cost of our algorithm makes feasible to use P LDel in real systems, instead of the Gabriel or the Relative Neighborhood graphs, which are not good spanners of U DG. ⋆ This work was partially supported by LaSIGE and by the FCT project INDIQoS POSI/CHS/41473/2001 via POSI and FEDER funds.

Shared Memory

Introduction to Reliable and Secure Distributed Programming, 2010

This chapter presents abstractions of shared memory. They represent distributed programming abstr... more This chapter presents abstractions of shared memory. They represent distributed programming abstractions, which are shared among processes and encapsulate data storage functionality accessible by read and write operations. The memory abstractions are called registers because they resemble those provided by multiprocessor machines at the hardware level, though in many cases, including in this chapter, they are implemented over processes that communicate by exchanging messages over a network and do not share ...

Basic Abstractions

Introduction to Reliable and Secure Distributed Programming, 2010

Applications that are deployed in practical distributed systems usually execute on a myriad of di... more Applications that are deployed in practical distributed systems usually execute on a myriad of different machines and communication infrastructures. Physical machines differ in the number of processors, type of processors, amount and speed of both volatile and persistent memory, and so on. Communication infrastructures differ in parameters such as latency, throughput, reliability, etc. On top of these machines and infrastructures, a huge variety of software components are sometimes needed to support one application: operating ...

Consensus

Introduction to Reliable and Secure Distributed Programming, 2010

This chapter is devoted to the consensus abstraction. The processes use consensus to agree on a c... more This chapter is devoted to the consensus abstraction. The processes use consensus to agree on a common value out of values they initially propose. Reaching consensus is one of the most fundamental problems in distributed computing. Any algorithm that helps multiple processes maintain common state or to decide on a future action, in a model where some processes may fail, involves solving a consensus problem.

Consensus Variants

Introduction to Reliable and Secure Distributed Programming, 2010

This chapter describes variants of the consensus abstraction introduced in the previous chapter. ... more This chapter describes variants of the consensus abstraction introduced in the previous chapter. These variants are motivated by applications of consensus to areas like fault-tolerant, replicated services, and distributed databases. In the variants we consider here, just like in consensus, the processes need to make consistent decisions, such as agreeing on one common value. However, most of the abstractions extend or change the interface of consensus, in order to satisfy the specific coordination requirements of an application. ...

Concluding Remarks

Introduction to Reliable and Secure Distributed Programming, 2010

Reliable Broadcast

Introduction to Reliable and Secure Distributed Programming, 2010

He said: "I could have been someone"; She replied: "So could anyone." (The Pogues)

Introduction

Introduction to Reliable and Secure Distributed Programming, 2010

Introduction to Reliable and Secure Distributed Programming

Dr. Christian Cachin IBM Research Z¨urich S¨aumerstrasse 4 Prof. Luıs Rodrigues INESC-ID Institut... more Dr. Christian Cachin IBM Research Z¨urich S¨aumerstrasse 4 Prof. Luıs Rodrigues INESC-ID Instituto Superior T´ecnico Rua Alves Redol 9 1000-029 Lisboa Portugal [email protected] 8803 R¨uschlikon Switzerland [email protected] Prof. Dr. Rachid Guerraoui Ecole Polytechnique F´edérale Lausanne (EPFL) Fac. Informatique et Communications Lab. Programmation Distribuée (LPD) Station 14 1015 Lausanne Bat. INR Switzerland [email protected] ISBN 978-3-642-15259-7 e-ISBN 978-3-642-15260-3 DOI 10.1007/978-3-642-15260-3 ...

A Generic Framework for Replicated Software Transactional Memories

2011 IEEE 10th International Symposium on Network Computing and Applications, 2011

Software Transactional Memory (STM) has emerged a powerful abstraction for managing access to sha... more Software Transactional Memory (STM) has emerged a powerful abstraction for managing access to shared data. Therefore, it is no surprise that a handful of different STM replication schemes have been proposed in the last recent years. In this context, we propose an architecture that facilitates the integration and execution of multiple replication techniques in a single, coherent, middleware infrastructure. This paves the way towards the development of autonomic mechanisms, able to select in runtime the most appropriate replication technique for the workload at hand.

Thicket: A Protocol for Building and Maintaining Multiple Trees in a P2P Overlay

2010 29th IEEE Symposium on Reliable Distributed Systems, 2010

One way to efficiently disseminate information in a P2P overlay is to rely on a spanning tree. Ho... more One way to efficiently disseminate information in a P2P overlay is to rely on a spanning tree. However, in a tree, interior nodes support a much higher load than leaf nodes. Also, the failure of a single node can break the tree, impairing the reliability of the dissemination protocol. These problems can be addressed by using multiple trees, such that each node is interior in just a few trees and a leaf node in the remaining; the multiple trees allow to achieve load distribution and also to send redundant information for fault-tolerance. This paper proposes Thicket, a decentralized algorithm to efficiently build and maintain such multiple trees over a single unstructured overlay network. The algorithm has been implemented and is extensively evaluated using simulation in a P2P overlay with 10.000 nodes.

Policies for Efficient Data Replication in P2P Systems

2013 International Conference on Parallel and Distributed Systems, 2013

This paper addresses the problem of maintaining replicated data in large scale P2P systems. Altho... more This paper addresses the problem of maintaining replicated data in large scale P2P systems. Although this topic has been extensively studied in the literature, to maintain replicated data in this setting, in an efficient manner, still remains a significant challenge. This paper proposes novel policies to address this problem and evaluates their performance against different criteria, such as monitoring costs, data transfer costs, and load unbalance costs. We show that one of these new policies significantly outperforms previous work. Interestingly, this policy is based on a somehow counter-intuitive approach, that uses less reliable nodes to store the most accessed data items. The insights to derive this policy were obtained from an in depth analysis of existing solutions, that is also captured in the paper.

Emergent Structure in Unstructured Epidemic Multicast

37th Annual IEEE/IFIP International Conference on Dependable Systems and Networks (DSN'07), 2007

In epidemic or gossip-based multicast protocols, each node simply relays each message to some ran... more In epidemic or gossip-based multicast protocols, each node simply relays each message to some random neighbors, such that all destinations receive it at least once with high probability. In sharp contrast, structured multicast protocols explicitly build and use a spanning tree to take advantage of efficient paths, and aim at having each message received exactly once. Unfortunately, when failures occur, the tree must be rebuilt. Gossiping thus provides simplicity and resilience at the expense of performance and resource efficiency.

Gossip-Based Broadcast

Handbook of Peer-to-Peer Networking, 2009

Gossip, or epidemic, protocols have emerged as a powerful strategy to implement highly scalable a... more Gossip, or epidemic, protocols have emerged as a powerful strategy to implement highly scalable and resilient reliable broadcast primitives on large scale peer-to-peer networks. Epidemic protocols are scalable because they distribute the load among all nodes in the system and resilient because they have an intrinsic level of redundancy that masks node and network failures. This chapter provides an introduction to gossip-based broadcast on largescale unstructured peer-to-peer overlay networks: it surveys the main results in the field, discusses techniques to build and maintain the overlays that support efficient dissemination strategies, and provides an in-depth discussion and experimental evaluation of two concrete protocols, named HyParView and Plumtree.

Large-Scale Peer-to-Peer Autonomic Monitoring

2008 IEEE Globecom Workshops, 2008

The increasing scale and complexity of distributed system motivates the need for autonomous manag... more The increasing scale and complexity of distributed system motivates the need for autonomous management. One of the key aspects in the management of distributed systems is the issue of component monitoring. Component monitoring is particularly challenging in large-scale dynamic systems, given the need to ensure that each component is monitored by at least one non-faulty component, despite joins, leaves, and failures, both at node and at network level. This paper proposes that components selforganize in an unstructured overlay network of constant degree in order to ensure that each component is always monitored by a threshold of other components.

Rollerchain

Proceedings of the Workshop on Posters and Demos Track - PDT '11, 2011

Abstract This paper presents Rollerchain, a novel Distributed Hash Table that offers high availab... more Abstract This paper presents Rollerchain, a novel Distributed Hash Table that offers high availability of stored data in an efficient manner through the combination of unstructured and structured overlay networks. The unstructured component maintains clusters of fully connected nodes, where each cluster acts as a virtual node in the structured component. We have implemented a prototype of Rollerchain that we have used to experimentally validate its better resilience and reduced costs when compared with other state of the art solutions.

Modelling adaptive services for distributed systems

Proceedings of the 2008 ACM symposium on Applied computing - SAC '08, 2008

There exists a growing class of distributed applications that require adaptive middleware service... more There exists a growing class of distributed applications that require adaptive middleware services, i.e., services that are able to monitor changes in the execution environment and in the user's requirements, reacting to these changes by adapting their behaviour. This paper proposes modelling primitives that allow to describe the adaptation logic of distributed applications that use reconfigurable service compositions.

Semantically Reliable Broadcast

Concurrency in Dependable Computing, 2002

Overnesia: an Overlay Network for Virtual Super-Peers

On Replication of Software Transactional Memories

Lecture Notes in Computer Science, 2008

Software Transactional Memory (STM) systems have garnered considerable interest of late due to th... more Software Transactional Memory (STM) systems have garnered considerable interest of late due to the recent architectural trend that has led to the pervasive adoption of multi-core CPUs. STMs represent an attractive solution to spare programmers from the pitfalls of conventional explicit lock-based thread synchronization, leveraging on concurrency-control concepts used for decades by the database community to simplify the mainstream parallel programming [1].

Fast Localized Delaunay Triangulation

Lecture Notes in Computer Science, 2005

A localized Delaunay triangulation owns the following interesting properties in a wireless ad hoc... more A localized Delaunay triangulation owns the following interesting properties in a wireless ad hoc setting: it can be built with localized information, the communication cost imposed by control information is limited and it supports geographical routing algorithms that offer guaranteed convergence. This paper presents a localized algorithm that builds a graph called planar localized Delaunay triangulation, P LDel, known to be a good spanner of the unit disk graph, U DG. Unlike previous work, our algorithm builds P LDel in a single communication step, maintaining a communication cost of O(n log n), which is within a constant of the optimum. This represents a significant practical improvement over previous algorithms with similar theoretical bounds. Furthermore, the small cost of our algorithm makes feasible to use P LDel in real systems, instead of the Gabriel or the Relative Neighborhood graphs, which are not good spanners of U DG. ⋆ This work was partially supported by LaSIGE and by the FCT project INDIQoS POSI/CHS/41473/2001 via POSI and FEDER funds.

Shared Memory

Introduction to Reliable and Secure Distributed Programming, 2010

This chapter presents abstractions of shared memory. They represent distributed programming abstr... more This chapter presents abstractions of shared memory. They represent distributed programming abstractions, which are shared among processes and encapsulate data storage functionality accessible by read and write operations. The memory abstractions are called registers because they resemble those provided by multiprocessor machines at the hardware level, though in many cases, including in this chapter, they are implemented over processes that communicate by exchanging messages over a network and do not share ...

Basic Abstractions

Introduction to Reliable and Secure Distributed Programming, 2010

Applications that are deployed in practical distributed systems usually execute on a myriad of di... more Applications that are deployed in practical distributed systems usually execute on a myriad of different machines and communication infrastructures. Physical machines differ in the number of processors, type of processors, amount and speed of both volatile and persistent memory, and so on. Communication infrastructures differ in parameters such as latency, throughput, reliability, etc. On top of these machines and infrastructures, a huge variety of software components are sometimes needed to support one application: operating ...

Consensus

Introduction to Reliable and Secure Distributed Programming, 2010

This chapter is devoted to the consensus abstraction. The processes use consensus to agree on a c... more This chapter is devoted to the consensus abstraction. The processes use consensus to agree on a common value out of values they initially propose. Reaching consensus is one of the most fundamental problems in distributed computing. Any algorithm that helps multiple processes maintain common state or to decide on a future action, in a model where some processes may fail, involves solving a consensus problem.

Consensus Variants

Introduction to Reliable and Secure Distributed Programming, 2010

This chapter describes variants of the consensus abstraction introduced in the previous chapter. ... more This chapter describes variants of the consensus abstraction introduced in the previous chapter. These variants are motivated by applications of consensus to areas like fault-tolerant, replicated services, and distributed databases. In the variants we consider here, just like in consensus, the processes need to make consistent decisions, such as agreeing on one common value. However, most of the abstractions extend or change the interface of consensus, in order to satisfy the specific coordination requirements of an application. ...

Concluding Remarks

Introduction to Reliable and Secure Distributed Programming, 2010

Reliable Broadcast

Introduction to Reliable and Secure Distributed Programming, 2010

He said: "I could have been someone"; She replied: "So could anyone." (The Pogues)

Introduction

Introduction to Reliable and Secure Distributed Programming, 2010

Introduction to Reliable and Secure Distributed Programming

Dr. Christian Cachin IBM Research Z¨urich S¨aumerstrasse 4 Prof. Luıs Rodrigues INESC-ID Institut... more Dr. Christian Cachin IBM Research Z¨urich S¨aumerstrasse 4 Prof. Luıs Rodrigues INESC-ID Instituto Superior T´ecnico Rua Alves Redol 9 1000-029 Lisboa Portugal [email protected] 8803 R¨uschlikon Switzerland [email protected] Prof. Dr. Rachid Guerraoui Ecole Polytechnique F´edérale Lausanne (EPFL) Fac. Informatique et Communications Lab. Programmation Distribuée (LPD) Station 14 1015 Lausanne Bat. INR Switzerland [email protected] ISBN 978-3-642-15259-7 e-ISBN 978-3-642-15260-3 DOI 10.1007/978-3-642-15260-3 ...

A Generic Framework for Replicated Software Transactional Memories

2011 IEEE 10th International Symposium on Network Computing and Applications, 2011

Software Transactional Memory (STM) has emerged a powerful abstraction for managing access to sha... more Software Transactional Memory (STM) has emerged a powerful abstraction for managing access to shared data. Therefore, it is no surprise that a handful of different STM replication schemes have been proposed in the last recent years. In this context, we propose an architecture that facilitates the integration and execution of multiple replication techniques in a single, coherent, middleware infrastructure. This paves the way towards the development of autonomic mechanisms, able to select in runtime the most appropriate replication technique for the workload at hand.

Thicket: A Protocol for Building and Maintaining Multiple Trees in a P2P Overlay

2010 29th IEEE Symposium on Reliable Distributed Systems, 2010

One way to efficiently disseminate information in a P2P overlay is to rely on a spanning tree. Ho... more One way to efficiently disseminate information in a P2P overlay is to rely on a spanning tree. However, in a tree, interior nodes support a much higher load than leaf nodes. Also, the failure of a single node can break the tree, impairing the reliability of the dissemination protocol. These problems can be addressed by using multiple trees, such that each node is interior in just a few trees and a leaf node in the remaining; the multiple trees allow to achieve load distribution and also to send redundant information for fault-tolerance. This paper proposes Thicket, a decentralized algorithm to efficiently build and maintain such multiple trees over a single unstructured overlay network. The algorithm has been implemented and is extensively evaluated using simulation in a P2P overlay with 10.000 nodes.

Policies for Efficient Data Replication in P2P Systems

2013 International Conference on Parallel and Distributed Systems, 2013

This paper addresses the problem of maintaining replicated data in large scale P2P systems. Altho... more This paper addresses the problem of maintaining replicated data in large scale P2P systems. Although this topic has been extensively studied in the literature, to maintain replicated data in this setting, in an efficient manner, still remains a significant challenge. This paper proposes novel policies to address this problem and evaluates their performance against different criteria, such as monitoring costs, data transfer costs, and load unbalance costs. We show that one of these new policies significantly outperforms previous work. Interestingly, this policy is based on a somehow counter-intuitive approach, that uses less reliable nodes to store the most accessed data items. The insights to derive this policy were obtained from an in depth analysis of existing solutions, that is also captured in the paper.

Emergent Structure in Unstructured Epidemic Multicast

37th Annual IEEE/IFIP International Conference on Dependable Systems and Networks (DSN'07), 2007

In epidemic or gossip-based multicast protocols, each node simply relays each message to some ran... more In epidemic or gossip-based multicast protocols, each node simply relays each message to some random neighbors, such that all destinations receive it at least once with high probability. In sharp contrast, structured multicast protocols explicitly build and use a spanning tree to take advantage of efficient paths, and aim at having each message received exactly once. Unfortunately, when failures occur, the tree must be rebuilt. Gossiping thus provides simplicity and resilience at the expense of performance and resource efficiency.

Gossip-Based Broadcast

Handbook of Peer-to-Peer Networking, 2009

Gossip, or epidemic, protocols have emerged as a powerful strategy to implement highly scalable a... more Gossip, or epidemic, protocols have emerged as a powerful strategy to implement highly scalable and resilient reliable broadcast primitives on large scale peer-to-peer networks. Epidemic protocols are scalable because they distribute the load among all nodes in the system and resilient because they have an intrinsic level of redundancy that masks node and network failures. This chapter provides an introduction to gossip-based broadcast on largescale unstructured peer-to-peer overlay networks: it surveys the main results in the field, discusses techniques to build and maintain the overlays that support efficient dissemination strategies, and provides an in-depth discussion and experimental evaluation of two concrete protocols, named HyParView and Plumtree.

Large-Scale Peer-to-Peer Autonomic Monitoring

2008 IEEE Globecom Workshops, 2008

The increasing scale and complexity of distributed system motivates the need for autonomous manag... more The increasing scale and complexity of distributed system motivates the need for autonomous management. One of the key aspects in the management of distributed systems is the issue of component monitoring. Component monitoring is particularly challenging in large-scale dynamic systems, given the need to ensure that each component is monitored by at least one non-faulty component, despite joins, leaves, and failures, both at node and at network level. This paper proposes that components selforganize in an unstructured overlay network of constant degree in order to ensure that each component is always monitored by a threshold of other components.

Rollerchain

Proceedings of the Workshop on Posters and Demos Track - PDT '11, 2011

Abstract This paper presents Rollerchain, a novel Distributed Hash Table that offers high availab... more Abstract This paper presents Rollerchain, a novel Distributed Hash Table that offers high availability of stored data in an efficient manner through the combination of unstructured and structured overlay networks. The unstructured component maintains clusters of fully connected nodes, where each cluster acts as a virtual node in the structured component. We have implemented a prototype of Rollerchain that we have used to experimentally validate its better resilience and reduced costs when compared with other state of the art solutions.

Modelling adaptive services for distributed systems

Proceedings of the 2008 ACM symposium on Applied computing - SAC '08, 2008

There exists a growing class of distributed applications that require adaptive middleware service... more There exists a growing class of distributed applications that require adaptive middleware services, i.e., services that are able to monitor changes in the execution environment and in the user's requirements, reacting to these changes by adapting their behaviour. This paper proposes modelling primitives that allow to describe the adaptation logic of distributed applications that use reconfigurable service compositions.

Semantically Reliable Broadcast

Concurrency in Dependable Computing, 2002

Overnesia: an Overlay Network for Virtual Super-Peers