Gossip-based Protocols Research Articles

Topological network features determine convergence rate of distributed average algorithms

Gossip algorithms are message-passing schemes designed to compute averages and other global functions over networks through asynchronous and randomised pairwise interactions. Gossip-based protocols have drawn much attention for achieving robust and fault-tolerant communication while maintaining simplicity and scalability. However, the frequent propagation of redundant information makes them inefficient and resource-intensive. Most previous works have been devoted to deriving performance bounds and developing faster algorithms tailored to specific structures. In contrast, this study focuses on characterising the effect of topological network features on performance so that faster convergence can be engineered by acting on the underlying network rather than the gossip algorithm. The numerical experiments identify the topological limiting factors, the most predictive graph metrics, and the most efficient algorithms for each graph family and for all graphs, providing guidelines for designing and maintaining resource-efficient networks. Regression analyses confirm the explanatory power of structural features and demonstrate the validity of the topological approach in performance estimation. Finally, the high predictive capabilities of local metrics and the possibility of computing them in a distributed manner and at a low computational cost inform the design and implementation of a novel distributed approach for predicting performance from the network topology.

Distributed cooperative learning over time-varying random networks using a gossip-based communication protocol

Motivated by applications of distributed estimation and distributed decision making in wireless sensor networks (WSNs) and unmanned aerial vehicle (UAV) networks, we study a distributed learning problem over time-varying undirected random networks. Using a gossip-based communication protocol, a novel distributed cooperative learning (DCL) algorithm, termed the gossip-based DCL (GBDCL) algorithm, is presented to solve the problem by training the raw data distributed and blocked throughout different nodes. Exploiting the robustness of the gossip-based protocol, each node is guaranteed to build the same learning model in theory against random disconnections and communication route variations in the network topology. It is proved that the GBDCL algorithm converges to the optimal consensus asymptotically. The correctness and effectiveness of the presented GBDCL algorithm are verified in the theoretical analysis and simulations.

Mining frequent items in unstructured P2P networks

Large scale decentralized systems, such as P2P, sensor or IoT device networks are becoming increasingly common, and require robust protocols to address the challenges posed by the distribution of data and the large number of peers belonging to the network. In this paper, we deal with the problem of mining frequent items in unstructured P2P networks. This problem, of practical importance, has many useful applications. We design P2PSS, a fully decentralized, gossip-based protocol for frequent items discovery, leveraging the Space-Saving algorithm. We formally prove the correctness and theoretical error bound. Extensive experimental results clearly show that P2PSS provides very good accuracy and scalability, also in the presence of highly dynamic P2P networks with churning. To the best of our knowledge, this is the first gossip-based distributed algorithm providing strong theoretical guarantees for both the Approximate Frequent Items Problem in Unstructured P2P Networks and for the frequency estimation of discovered frequent items.

Decentralized PMU-Assisted Power System State Estimation With Reduced Interarea Communication

This paper presents a decentralized approach to multiarea power system state estimation using a combination of conventional measurement devices and newer phasor measurement units (PMU). We employ a reduced-order approach to state estimation, wherein the PMU observable and unobservable components of the state vector are estimated separately in each control area. The estimation problem is solved by exchanging information through an improved gossip-based protocol, which takes into account the weak measurement coupling between control areas. A detailed analysis of the proposed protocol reveals that the amount of information exchange is always less than the naive gossip-based scheme. We show through simulations on the IEEE 118-bus test system that, in practice, the gossip iterations converge in one iteration regardless of the number of control areas. Our simulations also show that the marginal information exchange is also considerably lower when using the proposed method.

Context-Aware Gossip-Based Protocol for Internet of Things Applications.

This paper proposes a gossip-based protocol that utilises a multi-factor weighting function (MFWF) that takes several parameters into account: residual energy, Chebyshev distances to neighbouring nodes and the sink node, node density, and message priority. The effects of these parameters were examined to guide the customization of the weight function to effectively disseminate data to three types of IoT applications: critical, bandwidth-intensive, and energy-efficient applications. The performances of the three resulting MFWFs were assessed in comparison with the performances of the traditional gossiping protocol and the Fair Efficient Location-based Gossiping (FELGossiping) protocol in terms of end-to-end delay, network lifetime, rebroadcast nodes, and saved rebroadcasts. The experimental results demonstrated the proposed protocol’s ability to achieve a much shorter delay for critical IoT applications. For bandwidth-intensive IoT application, the proposed protocol was able to achieve a smaller percentage of rebroadcast nodes and an increased percentage of saved rebroadcasts, i.e., better bandwidth utilisation. The adapted MFWF for energy-efficient IoT application was able to improve the network lifetime compared to that of gossiping and FELGossiping. These results demonstrate the high level of flexibility of the proposed protocol with respect to network context and message priority.

Novel Multidimensional Models of Opinion Dynamics in Social Networks

Unlike many complex networks studied in the literature, social networks rarely exhibit unanimous behavior, or consensus. This requires a development of mathematical models that are sufficiently simple to be examined and capture, at the same time, the complex behavior of real social groups, where opinions and actions related to them may form clusters of different size. One such model, proposed by Friedkin and Johnsen, extends the idea of conventional consensus algorithm (also referred to as the iterative opinion pooling) to take into account the actors' prejudices, caused by some exogenous factors and leading to disagreement in the final opinions. In this paper, we offer a novel multidimensional extension, describing the evolution of the agents' opinions on several topics. Unlike the existing models, these topics are interdependent, and hence the opinions being formed on these topics are also mutually dependent. We rigorous examine stability properties of the proposed model, in particular, convergence of the agents' opinions. Although our model assumes synchronous communication among the agents, we show that the same final opinions may be reached "on average" via asynchronous gossip-based protocols.

Simulating cheated results acceptance rates for gossip-based volunteer computing

In this article, we present two different methods to estimate the dissemination rates of cheated results for decentralised distribution algorithms designed for volunteer computing networks. First, we define gossip-based protocols and present a short taxonomy that is based on data sizes for categorisation of distribution algorithms. We show three different distribution algorithms that are suitable for volunteer computing based on gossip-based protocols. Here, we additionally present an attacker model showing different attack possibilities on such algorithms. To minimise the impact of cheaters on a volunteer computing job, we introduce cheat detection mechanisms based on positive and negative verification of computation results. Finally, to minimise the amount of needed cheat detection computations, we either use simulations with the help of cellular automatons and a mathematical model to estimate the dissemination rates of cheated results. With the help of our methods, the cheat detection rate can be adjusted in a way that the speed up is maximal and the computation overhead for the cheat detection is minimal.

Simulating cheated results acceptance rates for gossip-based volunteer computing

In this article, we present two different methods to estimate the dissemination rates of cheated results for decentralised distribution algorithms designed for volunteer computing networks. First, we define gossip-based protocols and present a short taxonomy that is based on data sizes for categorisation of distribution algorithms. We show three different distribution algorithms that are suitable for volunteer computing based on gossip-based protocols. Here, we additionally present an attacker model showing different attack possibilities on such algorithms. To minimise the impact of cheaters on a volunteer computing job, we introduce cheat detection mechanisms based on positive and negative verification of computation results. Finally, to minimise the amount of needed cheat detection computations, we either use simulations with the help of cellular automatons and a mathematical model to estimate the dissemination rates of cheated results. With the help of our methods, the cheat detection rate can be adjusted in a way that the speed up is maximal and the computation overhead for the cheat detection is minimal.

A distributed data clustering algorithm in P2P networks

Clustering is one of the important data mining issues, especially for large and distributed data analysis. Distributed computing environments such as Peer-to-Peer (P2P) networks involve separated/scattered data sources, distributed among the peers. According to unpredictable growth and dynamic nature of P2P networks, data of peers are constantly changing. Due to the high volume of computing and communications and privacy concerns, processing of these types of data should be applied in a distributed way and without central management. Today, most applications of P2P systems focus on unstructured P2P systems. In unstructured P2P networks, spreading gossip is a simple and efficient method of communication, which can adapt to dynamic conditions in these networks. Recently, some algorithms with different pros and cons have been proposed for data clustering in P2P networks. In this paper, by combining a novel method for extracting the representative data, a gossip-based protocol and a new centralized clustering method, a Gossip Based Distributed Clustering algorithm for P2P networks called GBDC-P2P is proposed. The GBDC-P2P algorithm is suitable for data clustering in unstructured P2P networks and it adapts to the dynamic conditions of these networks. In the GBDC-P2P algorithm, peers perform data clustering operation with a distributed approach only through communications with their neighbours. The GBDC-P2P does not need to rely on a central server and it performs asynchronously. Evaluation results demonstrate the superior performance of the GBDC-P2P algorithm. Also, a comparative analysis with other well-established methods illustrates the efficiency of the proposed method.

MobiStore: A system for efficient mobile P2P data sharing

MobiStore is a P2P data store for decentralized mobile computing, designed to achieve high availability and load balance. As P2P platforms, mobile devices connected to the Internet through WiFi or cellular networks are different from wired devices in two main aspects: (1) higher churn due to mobility, weak wireless signals, or battery constraints, and (2) significant variability in bandwidth and latency based on the point of attachment. These problems affect the stored content availability and skew the content serving load over the peers. MobiStore structures the mobile P2P network into clusters of redundant peers. The topology uses both algorithmically-defined and random edges among the peers of different clusters. The routing information is updated using a gossip-based protocol. Thus, MobiStore achieves, with high probability, O(1) lookup operations despite high churn and link variability. Inside the clusters, all peers replicate the content, which improves the content availability. Furthermore, based on the current load, MobiStore dynamically changes the number of peers inside the clusters and routes content request to randomly selected peers. These two dynamic techniques along with using consistent hashing to map content to peers balance the load over the peers. While some of these techniques are well known, the main contribution is on the novel ways of applying them to design and implement an efficient mobile P2P data store. Simulation results show MobiStore achieves an availability, i.e., lookup success rate, between 12–48 % higher than two baseline systems built over the MR-Chord and Chord P2P protocols; and it reduces the latency up to 9 times compared to these protocols. Finally, the results show MobiStore adapts to churn and workload to evenly distribute the requests across clusters and peers better than both baseline solutions.

Heal Gossip: A Secure Overlay for Unstructured P2P Networks

Gossip-based protocols are an e cient mechanism for man-aging pure unstructured peer-to-peer (P2P) networks. Such protocols are Newscast, Cyclone, Lbpcast, etc. They have overcome from several di culties of such P2P random overlay connection. Such difficulties are randomness, high churn rate, very large unstructured distributed net-work, etc. But the performance of all gossip-based protocols have been completely vanished by presence of few malicious nodes. Since, non-detectable messages and behaviour of attackers are not leave them se-cure. These malicious nodes divide the overlay into several isolated clusters such as in Hub Attack or may be engaged non-malicious nodes in such a way that they are denying actual work such as in Denial of Ser-vice (DoS) Attacks. For securing unstructured P2P networks, there are some existing security protocols such as Secure Peer Sampling (SPSS), TooLate, S-Gossip etc. They are able to identify the malicious nodes and restrict them from gossiping. But restricting some malicious nodes on each node is not sufficient the security purpose of such epidemic over-lays. Especially in completely distributed networks, the malicious nodes may affect other non-malicious nodes although they have been already captured and restricted for gossip on others. In this regards, a new gossip mechanism is proposed, named HealGos-sip. It uses an additional property to inform captured malicious nodes on a node to all its neighbours. This process helps to identify and re-strict malicious nodes faster than other security mechanism. The propose mechanism relief the non-malicious nodes from the group of malicious nodes while performing detecting process. Hence, the propose protocol reduces the communication overhead as well as paralyses almost all malicious nodes within the network. For confusing among malicious and non-malicious nodes while detecting, a new variant of Hub attack is pro-posed and is called Hide and Seek (HnS) attack. It is able to miss-lead existing security protocols regarding the restriction of malicious nodes from gossip.

The people's cloud

Peer-to-peer cloud computing could free us from the tyranny of data centers. With the right software, geographically distributed hardware can provide a unified cloud-computing resource. Some cloud-computing providers put all their hardware eggs in one datacenter basket [blue]. Others employ multiple data centers networked together [orange]. The logical extension is a peer-to-peer cloud made of individual computers [yellow].The computers in some P2P networks resemble gossiping office workers. Gossip-based protocols allow information to flow reliably, even if some computers leave the system and break previously established links [orange lines]. Gossip-based protocols are used to maintain an unstructured peer-to-peer network of individual computers, some of which do the work of one customer [orange] while other combinations serve different customers [blue and yellow].

Performance analysis of ad hoc routing protocols based on selective forwarding node algorithms

Network resources are essential in resource critical environments as ad hoc wireless networks (AWNs) due to their dynamic topology and broadcasting of packets over the wireless medium without a central controller. In order to perform a routing algorithm with minimum resource consumption, an ideal ad hoc routing protocol should be designed for low consumed resources such as low power consumption. In this paper focus on ad hoc routing protocols, Optimized Link State Routing Protocol (OLSR) and gossip-based routing protocols, based on selective forwarding node algorithms that reduce the control message overheads. Thus, we present performance analytical results to thoroughly evaluate the routing protocols and demonstrate our Adaptive-Gossiping method advantages over Static-Gossiping routing and OLSR protocols in densely deployed networks. This paper is an extended version of ICISA 2013 conference paper [12].

Tracking freeriders in gossip-based content dissemination systems

Gossip-based protocols have proven very efficient for disseminating high-bandwidth content such as video streams in a peer-to-peer fashion. However, for the protocols to work, nodes are required to collaborate by devoting a fraction of their upload bandwidth, a scarce resource for some of them, to forward the content they receive to other nodes. Consequently, such protocols suffer from freeriding, a common phenomenon on the Internet, which consists in selfishly benefiting from the system without contributing its fair share. Due to the dynamic nature and the inherent randomness of gossip protocols and to the high scalability requirements of video streaming systems, detecting freeriders is a difficult challenge.This paper presents LiFTinG, the first protocol for detecting freeriders, including colluding ones, in gossip-based content dissemination systems with asymmetric data exchanges. In addition, LiFTinG is still able to detect freeriders when network coding, a widely used technique to improve the efficiency of content dissemination, is used. LiFTinG relies on nodes to track abnormal behavior by cross-checking the history of their previous interactions and exploits the fact that nodes pick neighbors at random to prevent colluding nodes from mutually covering up their bad actions.We present a methodology for setting the parameters of LiFTinG to their optimal value, based on a theoretical analysis and we quantify theoretically the performance of LiFTinG. We derive, based on simulations, the optimal strategy of freeriders by taking into account, through a utility function, the benefit of freeriding and the probability of being detected. In addition to these simulations, we report on the deployment of LiFTinG on PlanetLab. In a 300-node system, where a stream of 674kbps is broadcasted, LiFTinG incurs a maximum overhead of only 8% and provides good detection results: For instance, with 10% of freeriders decreasing their contribution by up to 30%, LiFTinG detects 86% of the freeriders after only 30s and wrongfully expels only a few honest nodes (most of them actually being buggy).

ReCREW: A Reliable Flash-Dissemination System

In this paper, we explore a new form of dissemination that arises in distributed, mission-critical applications called Flash Dissemination. This involves the rapid dissemination of rich information to a large number of recipients in a very short period of time. A key characteristic of Flash Dissemination is its unpredictability (e.g., natural hazards), but when invoked it must harness all possible resources to ensure timely delivery of information. Additionally, it must scale to a large number of recipients and perform efficiently in highly heterogeneous (data, network) and failure prone environments. We investigate a peer-based approach based on the simple principle of transferring dissemination load to information receivers using foundations from broadcast networks, gossip theory, and random networks. Gossip-based protocols are well known for being stateless, scalable, and fault-tolerant; however, their performance degrades as content size increases, because of the propagation of redundant gossip messages. In this paper, we propose Concurrent Random Expanding Walkers (CREW), a smart gossip protocol designed to maximize the speed of dissemination by transmitting data only as needed, and by exploiting both intra- and internode concurrency. CREW is designed to support both content and network heterogeneity and deal with transmission failures without sacrificing dissemination speed. We implemented CREW on top of a scalable middleware environment that allows for deployment across several platforms and developed optimizations without compromising on the stateless nature of CREW. We evaluated CREW empirically and compared it to optimized implementations of popular gossip and peer-based systems. Our experiments show that CREW significantly outperforms both traditional gossip and current large content dissemination systems while sustaining its performance in the presence of network errors.

Autonomous aggregation in location aware ad hoc wireless networks

ABSTRACTGossip‐based epidemic protocols are used to aggregate data in distributed systems. This fault‐tolerant approach neither require maintenance of any global network state nor knowledge of network structure. However, although gossip‐based aggregation algorithms scale well for graphs with good expansion, their efficiency for sparse graphs is unexamined. In this paper, we analyze the feasibility and efficiency of a gossip aggregation protocol in wireless networks with low expansion. We propose a modification of the existing aggregation algorithm for use in locality‐aware, sparse, and static wireless networks. Our protocol terminates autonomously, uses less bandwidth than the original version, and removes the need for the leader election process while counting network nodes. Aggregates are calculated only over nodes placed in the vicinity, and nodes communicate only with their immediate neighbors by using a wireless broadcast. We evaluate our approach by simulation on sparse, irregular graphs with low expansion for the simplified system model. Furthermore, we analytically assess the worst‐case convergence time of this protocol for sparse wireless networks and also for the simplified system model. The proposed protocol can be used in the monitoring and alarming applications in wireless networks. Copyright © 2012 John Wiley & Sons, Ltd.

RRG: redundancy reduced gossip protocol for real-time N-to-N dynamic group communication

Real-time group communication is an indispensable part of many interactive multimedia applications over the internet. In scenarios that involve large group sizes, sporadic sources, high user churns, and random network failures, gossip-based protocols can potentially provide advantages over structure-based group communication algorithms in ease of deployment, scalability, and resiliency against churns and failures. In this paper, we propose a novel protocol called Redundancy Reduced Gossip for real-time N-to-N group communication. We show that our proposed protocol can achieve a considerably lower traffic load than conventional push-based gossip protocols and conventional push-pull gossip protocols for the same probability of successful delivery, with higher performance gains in networks with smaller delays. We derive a mathematical model for estimating the frame non-delivery probability and the traffic load from overhead, and demonstrate the general correctness of the model by simulation. We implement a functioning prototype conferencing system using the proposed protocol, completed with functions including NTP synchronization, dynamic group size estimation, redundancy suppression, and other features needed for proper operation. We perform experiments over the campus network and PlanetLab, and the prototype system demonstrates the ability of our protocol to maintain robust performance in real-world network environments.

Extreme distributed systems: from large scale to complexity

Modern distributed systems can easily consist of hundreds of thousands of computers, ranging from high-end powerful machines to low-end resource-constrained wireless devices. We label them as “extreme distributed systems,” as they push scalability and complexity well beyond traditional scenarios. The extremeness of these systems is now requiring that we reconsider our methods and techniques for their development, and, indeed, we are already witnessing a shift in thinking. For example, Barroso and Holze [1] have made a case for a holistic design of a data center, which they essentially see as a single computer system. In their approach, the process of designing a data center is very similar to the way we have been designing processors: we need to take into account compute elements, data and control paths, storage, power sources, heating issues, and so on. As another example, groups from Lancaster University and INRIA/IRISA in Rennes are working on the integration of component-based software development with gossip-based protocols to combine structural and emergent approaches toward large-scale distributed system development [2]. It seems to be inevitable that we should concentrate more on fully decentralized solutions, as witnessed by, for example, peer-to-peer systems. Decentralized organizations often combine local decision-making with dissemination of information in order to improve the decision-making process, exemplified by many epidemic-based and other bio-inspired approaches. In this light, we are seeing much more than just ensuring that the constituents of a distributed system are properly placed, organized, and connected: the design of a distributed system is becoming fully integrated with

Pulp: An adaptive gossip-based dissemination protocol for multi-source message streams

Gossip-based protocols provide a simple, scalable, and robust way to disseminate messages in large-scale systems. In such protocols, messages are spread in an epidemic manner. Gossiping may take place between nodes using push, pull, or a combination. Push-based systems achieve reasonable latency and high resilience to failures but may impose an unnecessarily large redundancy and overhead on the system. At the other extreme, pull-based protocols impose a lower overhead on the network at the price of increased latencies. A few hybrid approaches have been proposed—typically pushing control messages and pulling data—to avoid the redundancy of high-volume content and single-source streams. Yet, to the best of our knowledge, no other system intermingles push and pull in a multiple-senders scenario, in such a way that data messages of one help in carrying control messages of the other and in adaptively adjusting its rate of operation, further reducing overall cost and improving both on delays and robustness. In this paper, we propose an efficient generic push-pull dissemination protocol, Pulp, which combines the best of both worlds. Pulp exploits the efficiency of push approaches, while limiting redundant messages and therefore imposing a low overhead, as pull protocols do. Pulp leverages the dissemination of multiple messages from diverse sources: by exploiting the push phase of messages to transmit information about other disseminations, Pulp enables an efficient pulling of other messages, which themselves help in turn with the dissemination of pending messages. We deployed Pulp on a cluster and on PlanetLab. Our results demonstrate that Pulp achieves an appealing trade-off between coverage, message redundancy, and propagation delay.

To reach consensus using uninorm aggregation operator: A gossip-based protocol

Gossip-based protocols for group communication have attractive scalability and reliability properties. This paper presents a gossip-based protocol that enables agents to reach a consensus based on a specific uninorm aggregation operator. We theoretically analyze the convergence, the speed and the randomness features of this protocol as well as its extensions. This model can be used to handle the uncertainty and the fast convergence characteristic of collective decision dynamics. The experimental results show that this protocol is efficient, scalable, and resilient against the failures under various network sizes, and topologies. © 2012 Wiley Periodicals, Inc.