"Today we are witnessing a dramatic shift toward a data-driven economy, where the ability to efficiently and timely analyze huge amounts of data marks the difference between industrial success stories and catastrophic failures. In this scenario Storm, an open source distributed realtime computation system, represents a disruptive technology that is quickly gaining the favor of big players like Twitter and Groupon. A Storm application is modeled as a topology, i.e. a graph where nodes are operators and edges represent data flows among such operators. A key aspect in tuning Storm performance lies in the strategy used to deploy a topology, i.e. how Storm schedules the execution of each topology component on the available computing infrastructure. In this paper we propose two advanced generic schedulers for Storm that provide improved performance for a wide range of application topologies. The first scheduler works offline by analyzing the topology structure and adapting the deployment to it; the second scheduler enhance the previous approach by continuously monitoring system performance and rescheduling the deployment at run-time to improve overall performance. Experimental results show that these algorithms can produce schedules that achieve significantly better performances compared to those produced by Storm's default scheduler."

"This paper studies the problem of answering aggregation queries, satisfying the interval validity semantics, in a distributed system prone to continuous arrival and departure of participants. The interval validity semantics states that the query answer must be calculated considering contributions of at least all processes that remained in the distributed system for the whole query duration. Satisfying this semantics in systems experiencing unbounded churn is impossible due to the lack of connectivity and path stability between processes. This paper presents a novel architecture, namely Virtual Tree, for building and maintaining a structured overlay network with guaranteed connectivity and path stability in settings characterized by bounded churn rate. The architecture includes a simple query answering algorithm that provides interval valid answers. The overlay network generated by the Virtual Tree architecture is a tree-shaped topology with virtual nodes constituted by clusters of processes and virtual links constituted by multiple communication links connecting processes located in adjacent virtual nodes. We formally prove a bound on the churn rate for interval valid queries in a distributed system where communication latencies are bounded by a constant unknown by processes. Finally, we carry out an extensive experimental evaluation that shows the degree of robustness of the overlay network generated by the virtual tree architecture under different churn rates."

"Applications based on event processing are often designed to continuously evaluate set of events defined by sliding time windows. Solutions employing long-running continuous queries executed in-memory show their limits in applications characterized by a staggering growth of available sources that continuously produce new events at high rates (e.g. intrusion detection systems and algorithmic trading). Problems arise due to the complexities in maintaining large amounts of events in memory for continuous elaboration, and due to the difficulties in managing at run-time the network of elaborating nodes. A batch approach to this kind of computation provides a viable solution for scenarios characterized by non frequent computations of very large time windows. In this paper we propose a model for batch processing in time window event computations that allows the definition of multiple metrics for performance optimization. These metrics specifically take into account the organization of input data to minimize its impact on computation latency. The model is then instantiated on Hadoop, a batch processing engine based on the MapReduce paradigm, and a set of strategies for efficiently arranging input data is described and evaluated."

"In this paper we analyze the coordinated port scan attack where a single adversary coordinates a Group of Attackers (GoA) in order to obtain information on a set of target networks. Such orchestration aims at avoiding Local Intrusion Detection Systems checks allowing each host of the GoA to send a very few number of probes to hosts of the target network. In order to detect this complex attack we propose a collaborative architecture where each target network deploys local sensors that send alarms to a collaborative layer. This, in turn, correlates this data with the aim of (i) identifying coordinated attacks while (ii) reducing false positive alarms and (iii) correctly separating GoAs that act concurrently on overlapping targets. Locally deployed sensors adopt graph-based clustering techniques over non-established TCP connections to generate alarms. The collaborative layer employs a similarity approach to aggregate alarms and approximated optimization algorithms to separate distinct GoAs. The soundness of our approach is tested on real network traces. Tests show that collaboration among networks domains is mandatory to achieve accurate detection of coordinated attacks and sharp separation between GoAs that execute concurrent attacks on the same targets."

"The SIENA publish/subscribe system represents a proto- typical design for a distributed event notification service implementing the content-based publish/subscribe communication paradigm. A clear shortcoming of SIENA is represented by its static configuration that must be managed and updated by human administrators every time one of its internal processes (brokers) needs to be added or repaired (e.g. due to a crash failure). This problem limits the applicability of SIENA in large complex critical infrastructures where self-adaptation and -configuration are crucial requirements. In this paper we propose HSIENA, a hybrid architecture that complements SIENA by adding the ability to self-reconfigure after broker additions and removals. The architecture has a novel design that mixes the classic SIENA’s distributed architecture with a highly available cloud-based storage service."

"A distributed event notication service (ENS) is a middleware ar- chitecture commonly used to provide applications with scalable and ro- bust publish/subscribe communication primitives. A distributed ENS can route events toward subscribers using multiple paths with dier- ent lengths and latencies; as a consequence, subscribers can receive events out of order. In this paper, we propose a novel solution for out-of-order notication detection on top of an existing topic-based ENS. Our solution guarantees that events published on dierent topics will be either delivered in the same order to all the subscribers of those topics or tagged as out-of-order. The proposed algorithm is completely distributed and is able to scale with the system size while imposing a reasonable cost in terms of notication latency. Our solution improves the current state of the art solutions by dynamically handling subscrip- tions/unsubscriptions and by automatically adapting with respect to topic popularity changes."

"In the last few years telco providers are striving to migrate their services from the traditional Public Switch Telephone Network (PSTN) to so called Next Generation Networks (NGNs) based on standard IP connectivity. This switch is expected to produce a cost degression of 50% for CAPital EXpenditure (CAPEX), while OPerating EXpences (OPEX) remains fairly stable due to network management and energy costs. At the same time, the instantiation of new telco services (Voice over IP (VoIP), video conferencing, etc.) and the support of third party applications (such as support to smartphone applications, etc.) are expected to produce a big increase of the load of a telco provider at the core level. The goal of this work is to show how management and energy costs can be effectively reduced by leveraging autonomic approaches to move some NGN services toward the telco network edge while still providing Quality of Service (QoS) levels comparable with those provided by a traditional fully-managed infrastructure. This is done by taking into consideration the increase of the load of such services that is expected to raise by one order of magnitude in the near future. Specifically, we propose a hybrid architecture letting telco administrators reduce the number of servers in the provider managed network by exploiting home devices in the computation and by organizing them in a self-configuring Peer to Peer (P2P) system; in this way it is possible to reduce the overall system and operational costs. Our claims are supported by an experimental study based on both simulations and theoretical models that analyze the trade-off between the number of servers and home devices in order to guarantee a service within QoS constraints. Experiments are carried out on a realistic model that abstracts the lookup procedures within the NGN of a big telco provider (i.e., finding the IP address of a given unique user profile)."

"This paper studies the complexity of answering aggregation queries with the interval validity semantics in a distributed system prone to churn. The interval validity semantics states that the query answer must contain contributions from at least all the processes that were part of the system for the whole query execution time. Solving this problem in unstructured distributed systems with unbounded churn is impossible due to the lack of connectivity and path stability between processes. This paper presents a novel overlay management protocol, namely Virtual Tree OMP, that, under the assumption of bounded churn, maintains a structured overlay network with guaranteed con- nectivity and path stability. The resulting Virtual Tree graph is a tree-shaped topology with virtual nodes constituted by cliques of pro- cesses and virtual links constituted by multiple communication links connecting processes located in adjacent virtual nodes. This paper also introduces a distributed query algorithm that, working on top of the Virtual Tree graph, guarantees answers complying with the interval validity semantics. Finally, the paper explores through an extensive experimental analysis the performance characteristics of the overlay maintenance protocol under different churn levels."

"Home automation represents a growing market in the industrialized world. Today's systems are mainly based on ad hoc and proprietary solutions, with little to no interoperability and smart integration. However, in a not so distant future, our homes will be equipped with many sensors, actuators and devices, which will collectively expose services, able to smartly interact and integrate, in order to offer complex services providing even richer functionalities. In this paper we present the approach and results of SM4All -- Smart hoMes for All, a project investigating automatic services composition and advanced user interfaces applied to domotics."

"The increasing use of the Internet and the improvement in hardware technology led most of the application previously deployed in \closed" environments, such as business intelligence, smart environments, complex system software management, to federate into geographically-distributed systems. Such applications use \sense-and-respond" capabilities, i.e., they correlate basic events that could potentially occur at dierent sources and detect complex event patterns, in order to timely and properly react to changes that may happen within the system. In this context, a fundamen- tal a fundamental role is played by the data dissemination service that brings events from producers to consumers where complex event patterns are detected. In this paper we discuss the characteristics that a data dis- semination service should have in order to support in the best way the complex event pattern detection functionality, and present an assessment of a number of technologies that can be used to disseminate data in the earlier mentioned context. We also describe how those technologies can be eectively deployed in scenarios where numerous independent data sources produce large amounts of events in the form of high-throughput streams. Finally, we present a matching between distributed application require- ments and the capabilities oered by the data dissemination services used to implement them, highlighting which aspects should be considered in the design of novel middleware solutions to ll this gap."

"Choosing the best platform for distributed publish/subscribe communications usually implies a careful choice of the desired event selection model; content-based systems allow fine grained interest selection through expressive subscriptions usually at the cost of limited scalability; topic-based systems offer high-performance and scalable event dissemination at the cost of reduced expressiveness for the subscribers that must then locally filter uninteresting events. In this paper we present a novel solution for implementing a content-based event selection model on top of an existing topic-based system; our solution is based on the idea of dynamically partition the event space and mapping subscribers interests on the resulting discrete intervals represented by topics; the partitioning is automatically built and continuously adapted at runtime in a distributed way with the goal of maintaining predefined minimum target performance despite possible changes in the subscriber interests. The effectiveness of our solution is shown through an extensive set of experiments."

"Let consider a set of anonymous moving objects to be tracked in a binary sensor network. This paper studies the problem of associating deterministically a track revealed by the sensor network with the trajectory of an unique anonymous object, namely the Multiple Object Tracking and Identification (MOTI) problem. In our model, the network is represented by a sparse connected graph where each vertex represents a binary sensor and there is an edge between two sensors if an object can pass from one sensed region to another one without activating any other sensor. The di􏰁culty of MOTI lies in the fact that the trajectories of two or more objects can be so close that the corresponding tracks on the sensor network can no longer be distinguished (track merging), thus confusing the deterministic association between an object trajectory and a track. The paper presents several results. Worst show that MOTI cannot be solved on a general graph of ideal binary sensors even by an omniscient external observer if all the objects can freely move on the graph. Then, we describe some restrictions that can be imposed a priori either on the graph, on the object movements or both, to make the MOTI problem always solvable. In the case of absence of an omniscient observer, we show how our results can lead to the definition of distributed algorithms that are able to detect when the system is in a state where MOTI becomes unsolvable."

"Telco providers are in the phase of migrating their services from PSTN to so called Next Generation Networks (NGNs) based on standard IP connectivity. This switch is expected to produce a cost degression of 50% for CAPEX, while OPEX remains fairly stable due to network management and energy costs. At the same time we are expecting a big increase of the load of a telco provider at the core level due to the istantiation of new telco services (VoIP, video conferencing etc) and to the support of third parties services (such as support to smartphone applications, etc.). The goal of this work is to show how management and energy costs can be effectively reduced by leveraging autonomic approaches to move some NGN services toward the telco network edge while still providing QoS levels comparable with those provided by a traditional fully-managed infrastructure."

"Up to now Service Oriented Architectures and Event Driven Architectures have been considered as competing parties striving to conquer the crown of the standard paradigm for the implementation of complex distributed applications. Todays we are witnesses of large efforts to merge both paradigms and give birth to a new generation of middleware platforms that will inherit the best of both worlds. In this paper we describe how this marriage could be leveraged in order to design new dependable software systems. "

"Providing independent uniform samples from a system population poses considerable problems in highly dynamic settings, like P2P systems, where the number of participants and their unpredictable behavior (e.g., churn, crashes etc.) may introduce relevant bias. Current implementations of the Peer Sampling Service are designed to provide uniform samples only in static settings and do not consider that biased samples can directly affect the correctness of algorithms relying on a uniformity property or be exploited by a malicious adversary to increase the effectiveness of its attacks to the system. In this paper we provide a practical solution to the biasing problem by deploying a fully distributed Peer Sampling Correction Module on top of a given, possibly biased, peer sampling service. Samples provided by the peer sampling service will be locally processed by this module, using computationally efficient hashing functions, before getting to the application. The effectiveness of our approach is evaluated through an extensive simulation-based study. Finally, we show the efficiency of the Peer Sampling Correction Module in a case study, namely the target selection attack prevention."

"Let us consider a large scale distributed system and a query executed on top of it where every process has to contribute to the result. Informally, a query satisfies the interval validity property if its result has been calculated by retrieving data from a set of processes containing at least all those ones that have been present in the system during the whole query lifetime. If the system is prone to churn, it is easy to show that a query cannot deterministically satisfy interval validity. In this paper we propose a novel algorithm that can be used to support distributed queries by increasing the probability of a query to satisfy interval validity. The algorithm strives to (i) reduce the query calculation time (to reduce the net effect of churn) and to (ii) increase the robustness of the overlay network it builds by clustering nodes into cliques of limited size in order for their implementation to be still practical. The paper provides a set of experiments that show the tradeoff between the churn rate and the number of times the interval validity is satisfied."

"Home automation represents a growing market in the industrialized world. Today’s systems are mainly based on ad hoc and proprietary solution, with little to no in- teroperability and smart integration. However, in a not so distant future, devices installed in our home will be able to smartly interact and integrate in order to of- fer complex services with rich functionalities. Realiz- ing this kind of integration push developers to increase the amount of abstraction within the software architec- ture. In this paper we give a high-level view of what are the inherent trade-offs that stem from this process of abstraction and suggest how they could be tackled in these complex home automation systems. More specifi- cally we focus our analysis on two problems: concurrent execultion of multiple plans and failure detection."

"This paper studies the problem of realizing a common software clock among a large set of nodes without an external time reference (i.e., internal clock synchronization), any centralized control and where nodes can join and leave the distributed system at their will. The paper proposes an internal clock synchronization algorithm which combines the gossip-based paradigm with a nature-inspired approach, coming from the coupled oscillators phenomenon, to cope with scale and churn. The algorithm works on the top of an overlay network and uses a uniform peer sampling service to fullfill each node's local view. Therefore, differently from clock synchronization protocols for small scale and static distributed systems, here each node synchronizes regularly with only the neighbors in its local view and not with the whole system. Theoretical and empirical evaluations of the convergence speed and of the synchronization error of the coupled-based internal clock synchronization algorithm have been carried out, showing how convergence time and the synchronization error depends on the coupling factor and on the local view size. Moreover the variation of the synchronization error with respect to churn and the impact of a sudden variation of the number of nodes have been analyzed to show the stability of the algorithm. In all these contexts, the algorithm shows nice performance and very good self-organizing properties. Finally, we showed how the assumption on the existence of a uniform peer-sampling service is instrumental for the good behavior of the algorithm."

"Most distributed applications today receive events, process them and in turn create new events which are sent to other processes. Business intelligence, air traffic control, collaborative security, complex system software management are examples of such applications. In these applications basic events, potentially occurred at different sites, are correlated in order to detect complex event patterns formed by basic events that could have temporal and spatial relationships among them. In this context, a fundamental functionality is the data dissemination that brings events from event producers to event consumers where complex event patterns are detected. In this paper we discuss the characteristics that a Data Dissemination service should have in order to support in the best way the complex event pattern detection functionality. We consider event traffic can reach thousands of events per second coming from different event sources; that is, the data dissemination service has to sustain high throughput. Finally, we present an assessment of a number of technologies that can be used to disseminate data in the earlier mentioned context, discussing scenarios where those technologies can be effectively deployed."

"Smart home applications are currently implemented by vendor-specific systems managing mainly a few number of homogeneous sensors and actuators. However, the sharp increase of the number of intelligent devices in a house and the foreseen explosion of the smart home application market will change completely this vendor centric scenario towards open, expandable systems made up of a large number of cheap heterogeneous devices. As a matter of fact, new smart home solutions have to be able to takle with scalability, dynamicity and heterogeneity requirements. In this paper we present the architecture of a basic building block, namely a distributed repository service, for smart home systems. The repository stores data from heterogeneous devices deployed in the house that can be then retrieved by context aware applications implementing some home automation functionalities. Our architecture, based on a DHT, offers a completely decentralized and reliable storage service able to offer complex query functionalities."

"In industry practice, bilateral agreements are established between providers and consumers of services in order to regulate their business relationships. In particular, Quality of Service (QoS) requirements are specified in those agreements in the form of legally binding contracts named Service Level Agreements (SLA). Meeting SLAs allows providers to be seen in the eyes of their clients, credible, reliable, and trustworthy. This contributes to augment their reputation that can be considered an important and competitive advantage for creating potentially new business opportunities. In this paper we describe the design and evaluation of a framework that can be used for dynamically computing the reputation of a service provider as specified into SLAs. Specifically, our framework evaluates the reputation by taking into account two principal factors; namely, the run time behaviors of the providers within a specific service provision relationship, and the reputation of the providers in the rest of the system. A feedback-based approach is used to assess these two factors: through it consumers express a vote on the providers’ behavior according to the actions the providers undertake. We have carried out an evaluation that aimed at showing the feasibility of our approach. This paper discusses the principal results we have obtained by this evaluation."

"In the last few years there has been a growing interest in small, low-power hardware platforms that integrate sensing, processing and wireless communication capabilities that can be adopted to quickly deploy powerful wireless sensor networks. Applications for WSNs often apply the event-based interaction paradigm for communication among participants. In this paper we report our experiences with the testing of JORAM, a well known JMS implementation, on top of the Sun SPOT wireless sensor platform."

"Random walks can be conveniently exploited for implementing probabilistic algorithms to solve many searching problems arose by distributed applications, e.g., service discovery, p2p file sharing, etc. In this paper we consider random walks executed on uniform wireless networks and study how to reduce the expected number of walk steps required to reach a target, namely the hitting time. The latter is the main search performance metric of a random walk based algorithm, since it determines the average response to a search as well as its cost; thus, the actual convenience of using random walks compared to other solutions depends of achieving a low hitting time. We show how in uniform wireless networks, the natural implementation of a random walk which selects the next node to visit at random among all neighbors is not a good choice, since it has a strong negative effect on the hitting time. This paper studies such a negative effect analytically and proposes two neighbor selection rules aiming at reducing the hitting time. A simulation study confirms the benefits of the proposed solutions."

"Recent evidence of successful Internet-based attacks and frauds involving financial institutions highlights the inadequacy of the existing protection mechanisms, in which each institution implements its own isolated monitoring and reaction strategy. Analyzing on-line activity and detecting attacks on a large scale is an open issue due to the huge amounts of events that should be collected and processed. In this paper, we propose a large-scale distributed event processing system, called intelligence cloud, allowing the financial entities to participate in a widely distributed monitoring and detection effort through the exchange and processing of information locally available at each participating site. We expect this approach to be able to handle large amounts of events arriving at high rates from multiple domains of the financial scenario. We describe a framework based on the intelligence cloud where each participant can receive early alerts enabling them to deploy proactive countermeasures and mitigation strategies."

"Distributed event routing has emerged as a key technology for achieving scalable information dissemination. In particular it has been used as preferential communication backbone within publish/subscribe communication system. Its aim is to reduce the network and computational overhead per event diffusion to a set of interested recipients. This chapter introduces the reader to modern publish/subscribe systems through an overview on current techniques for event dissemination. The approach we follow proposes a decomposition of these architectures in functional layers. We survey current algorithms for event based routing, and possible overlay infrastructures in wired and mobile systems."

"This paper studies the existence and the regularity of Logarithmic Harary Graphs (LHGs). This study is motivated by the fact that these graphs are employed for modeling the communication topology to support efficient flooding in the presence of link and node failures when considering an initial arbitrary number of nodes n. Therefore, the capability to identify graph constraints that allow the construction of LHGs for the largest number of pairs (n,k) (where k is the desired degree of connectivity to be tolerant to failures) becomes of primary importance. The paper presents several results in that direction. We introduce a graph constraint, namely K-PASTED-TREE, that allows the construction of a LHG for every pair (n,k) such that n≥2k. Secondly we present another graph constraint for LHG, namely K-DIAMOND, which is equivalent to K-PASTED-TREE in terms of capability to construct LHGs for any pair (n,k). The interest of K-DIAMOND lies in the fact that, for a given k, K-DIAMOND allows us to construct more regular graphs than K-PASTED-TREE does. A k-regular graph shows the minimal number of links required by a k-connected graph, leading to minimal flooding cost. The paper formally shows, in particular, that there are an infinite number of pairs (n,k), such that there exists a k-regular LHG for the pair (n,k) that satisfies K-DIAMOND and does not satisfy K-PASTED-TREE."

"In this work we present a novel abstraction that allows a set of distributed processes, aware of their respective positions in space, to collectively maintain information associated to an area in the physical world. This abstraction is a logical ob ject shared between participating processes and provides two operations, namely read and write, as for traditional registers. This paper provides a specification and implementation of such a shared ob ject for non concurrent writes and multiple readers, then describes a possible extension to concurrent writes. We also provide insights on how this kind of ob jects could be useful to program location-aware applications. "

"This paper proposes an internal clock synchronizationalgorithm for very large number of processes that is ableto (i) self-synchronize their local clocks without any centralcontrol and (ii) resist to attacks of an adversary whose aimis to put out-of-synchronization as many correct processesas possible. To cope with scale the algorithm utilizes thegossip-based paradigm where each process has a limitedview of the system, while to resist to attacks the algorithmemploys a filtering mechanism based on the notion of α-trimmed mean to filter out out-of-range clock values. Thealgorithm shows nice convergence in presence of networkserrors and in absence of the adversary. When the adversarytakes control of some of the processes in the system, we define two goals for the adversary, actually two predicates, tomeasure the strength of the attack. The first one capturesthe percentage of time in which at least one correct is out ofsynchronization and the second one when all correct processes are out of synchronization. The paper presents anextensive simulation study showing under which conditions(in terms of number of corrupted processes and size of localviews) these two goals can be achieved by the adversary.Interestingly, these results can be exploited by applicationsthat can tolerate either a certain time in which some correct process is non-synchronized or a certain percentage ofcorrect processes that is non-synchronized."

"This paper studies the existence and the regularity of Logarithmic Harary Graphs (LHGs). This study is motivated by the fact that these graphs are employed for modeling the communication topology to support efficient flooding in presence of link and node failures when considering an initial arbitrary number of nodes n. Therefore, the capability to identify graph constraints that allow the construction of LHGs for the largest number of pairs (n, k) (where k is the desired degree of connectivity to be tolerant to failures) becomes of primary importance. The paper presents several results in that direction. We introduce a graph constraint, namely K-TREE, that allows the construction of a LHG for every pair (n, k) such that n ges 2k. Secondly we presents another graph constraint for LHG, namely KDIAMOND, which is equivalent to K-TREE in terms of capability to construct LHGs for any pair (n, k). The interest of K-DIAMOND lies in the fact that, for a given k, KDIAMOND allows to construct more regular graphs than K-TREE does. A k-regular graph shows the minimal number of links required by a k-connected graph, leading tominimal flooding cost. The paper formally shows, in particular, that there are an infinite number of pairs (n, k), such that there exists a k-regular LHG for the pair (n, k) that satisfies K-DIAMOND and does not satisfy K-TREE."

"Middleware for timely and reliable data dissemination is a fundamental building block of the Event Driven Architecture (EDA) which is the ideal platform for developing a large class of sense-and-react applications such as air traffic control, defense systems, stock exchange etc. Many of these middlewares are compliant to the Data Distribution Service specification and, due to the nature of the application domain, they have been traditionally designed to be deployed on strictly controlled and managed environments where they show predictable behaviors and performance. However, several very recent studies show that EDA can boost efficiency of an enterprise from several perspectives (e.g., logistic, finance, stock management). However, the enterprise setting is far from being managed as it can be characterized, for example, by geographic inter-domain scale and heterogeneous resources. In this paper we present a study aimed at assessing the strengths and weaknesses of a commercial Data Distribution Service implementation deployed on an unmanaged setting. Our experiments campaign outlines that, if the application manages a small number of homogeneous resources, this middleware can perform timely and reliably as long as there is no event fragmentation at the network transport level. In a more general setting with fragmentation and heterogeneous resources (typical of the enterprise scenarios), reliability and timeliness rapidly degenerate pointing out a clear need of research in the field of self-configuring scalable event dissemination with QoS guarantee on unmanaged settings."

"Peer-to-peer systems (P2P) have become a popular technique to design large-scale distributed applications in unmanaged inter-domain settings, such as file sharing or chat systems, thanks to their capabilities to self-organize and evenly split the load among peers. Recently, enterprises owning a large IT hardware and software infrastructure started looking at these P2P technologies as a mean both to reduce costs and to help their technical divisions to manage huge number of devices characterized by a high level of cooperation and a relatively low churn. Gaining a quick exclusive access to the system for maintenance or auditing purposes in these enterprise infrastructures is a fundamental operation to be implemented. Conversely, this kind of operation is usually not an issue in the previously mentioned inter-domain setting, where peers are inherently independent and cannot be managed. In the context of classical distributed applications, quorum systems have been considered as a major building block for implementing many paradigms, from distributed mutual exclusion to data replication management. In this paper, we explore how to architect decentralized protocols implementing quorum systems in Distributed Hash Table based cooperative P2P networks. Our results show that quorum systems taken ``as is'' from the literature and directly applied to such networks are not scalable due to the high load imposed onto the underlying network. This paper introduces some design principles for both quorum systems and protocols using them that boost their scalability and performance. These design principles consist in a dynamic and decentralized selection of quorums and in the exposition and exploitation of internals of the DHT. As a third design principle it is also shown how to redesign quorum systems to enable efficient decentralization. We show that by combining these design principles in a cooperative environment with relatively low churn it is possible to minimize the imposed load in the system, in terms of sites contacted to obtain a quorum, and the latency of quorum acquisition. "

"Clock synchronization is a fundamental building block for many distributed applications. As such, the topic has been widely studied for many years, and several algorithms exist which address different scales, ranging from local area networks (LAN), to wide area networks (WAN). Network Time Protocol (NTP) [21, 22], has emerged as a standard de facto for external clock synchronization in both LAN and WAN settings. However there exists an emergent class of applications and services, operating in very challenging settings, for which the problem synchronizing clocks is far from being solved. These applications have to operate without any assumption on deployed functionalities, pre-existing infrastructure, or centralized control, while being able to tolerate network dynamism, due to crashes or to node joining or leaving the system, and scaling from few hundred to tens of thousands of nodes. For instance, publish/subscribe middleware, such as the data distribution service [10] requires synchronized clocks, however in several relevant scenarios, due to security issues, or limited assumptions on the infrastructure, it cannot assume that members of the system, either have access to an NTP server, or are equipped with an NTP daemon. "

"Recently, enterprises owning a large IT hardware and software infrastructure have started looking at Peer-to-peer technologies as a mean both to reduce costs and to help their technical divisions to manage huge number of devices characterized by a high level of cooperation and a relatively low churn. Obtaining the complete and exclusive control of the system for maintenance or auditing purposes in these enterprise infrastructures is a fundamental operation to be implemented. In the context of classical distributed applications, quorum systems have been considered as a major building block for implementing many paradigms, from distributed mutual exclusion to data replication management. In this paper, we explore how to architect decentralized protocols implementing quorum systems in Distributed Hash Table based cooperative P2P networks. This paper introduces some design principles for both quorum systems and protocols using them that boost their scalability and performance. These design principles consist of a dynamic and decentralized selection of quorums and in the exposure and exploitation of internals of the DHT. As a third design principle it is also shown how to redesign quorum systems to enable efficient decentralization."

"The publish/subscribe interaction paradigm is today becoming mainstream in a large number of very large scale applications like news syndication (with RSS) or massive multiplayer games. These applications are often still implemented by means of centralized services that will hardly scale with the user growth expected in the next years. Modern publish/subscribe systems are striving to address these scalability needs to play a dominant role in this future market. A very important contribution, on the road to reach this goal, is given by the interest clustering techniques adopted by these systems. Interest clustering aims at putting in close applicative relationship groups of users sharing similar interests in order to reduce the effort needed to dispatch a message to group. This technique can be applied to event dissemination mechanisms based on filtering to reduce the total amount of messages generated during event routing and, consequently, improve the overall system performance. In this paper we explore this topic to discover the potentialities of interest clustering, to understand how it can be implemented in a publish/subscribe system, and to study, through a small focussed survey, the central role played by this technique in modern systems."

"This paper studies the problem of associating deterministically a track revealed by a binary sensor network with the trajectory of a unique moving anonymous object, namely the Multiple Object Tracking and Identification (MOTI) problem. In our model, the network is represented by a sparse connected graph where each vertex represents a binary sensor and there is an edge between two sensors if an object can pass from a sensed region to another without activating any other remaining sensor. The difficulty of MOTI lies in the fact that trajectories of two or more objects can be so close (track merging) that the corresponding tracks on the sensor network can no longer be distinguished, thus confusing the deterministic association between an object trajectory and a track. The paper presents several results. We first show that MOTI cannot be solved on a general graph of ideal binary sensors even by an omniscient external observer if all the objects can freely move on the graph. Then, we describe some restrictions that can be imposed a priori either on the graph, on the object movements or both, to make MOTI problem always solvable. We also discuss the consequences of our results and present some related open problems."

"In this paper, we describe the aims and preliminary results of FADA, a framework for developing resilient services and reasoning on mobile systems. "

"This paper proposes an internal clock synchronization algorithm which combines the gossip-based paradigm with a nature-inspired approach coming from the coupled oscillators phenomenon. The proposed solution allows a very large number of clocks to self-synchronize without any central control, despite node departure and arrival. This addresses the needs of an emergent class of large-scale peer-to-peer applications that have to operate without any assumptions on the underlying infrastructure. Empirical evaluation shows extremely good convergence and stability under different network settings."

"In recent years the labels "gossip" and "gossip-based" have been applied to an increasingly general class of algorithms, including approaches to information aggregation, overlay network management and clock synchronization. These algorithms are intuitively similar, irrespective of their purpose. Their distinctive features include relying on local information, being round-based and relatively simple, and having a bounded information transmission and processing complexity in each round. Our position is that this class can and should be significantly extended to involve algorithms from other disciplines that share the same or similar distinctive features, like certain parallel numerical algorithms, routing protocols, bio-inspired algorithms and cellular automata, to name but a few. Such a broader perspective would allow us to import knowledge and tools to design and understand gossip-based distributed systems, and we could also export accumulated knowledge to re-interpret some of the problems in other disciplines, such as vehicular traffic control. In this position paper we describe a number of areas that show parallels with gossip protocols. These example areas will hopefully serve as inspiration for future research. In addition, we believe that comparisons with other fields also helps clarify the definition of gossip protocols and represents a necessary first step towards an eventual formal definition."

"Recently many scalable and efficient solutions for event dissemination in publish/subscribe (pub/sub) systems have appeared in the literature. This dissemination is usually done over an overlay network of brokers and its cost can be measured as the number of messages sent over the overlay to allow the event to reach all intended subscribers. Efficient solutions to this problem are often obtained through smart dissemination algorithms that avoid flooding events on the overlay. In this paper we propose a complementary approach, that is obtaining efficient event dissemination by reorganizing the overlay network topology. More specifically, this reorganization is done through a self-organizing algorithm executed by brokers whose aim is to directly connect, through overlay links, pairs of brokers matching same events. In this way, on average, the number of brokers involved in an event dissemination decreases, thus, reducing its cost. Even though the paradigm of the self-organizing algorithm is general and then applicable to any overlay-based pub/sub system, its concrete implementation depends on the specific system. As a consequence we studied the effect of the introduction of the self-organizing algorithm in the context of a specific system implementing a tree-based routing strategy, namely SIENA, showing the actual performance benefits through an extensive simulation study. In particular performance results point out the capacity of the algorithm to converge to an overlay topology accommodating efficient event dissemination with respect to a specific scenario. Moreover, the algorithm shows a significant capacity to adapt the overlay network topology to continuously changing scenarios while keeping an efficient behavior with respect to event dissemination."

"The completely decoupled interaction model offered by the publish/subscribe communication paradigm perfectly suits the interoperability needs of todays large-scale, dynamic, peer-to-peer applications. Unmanaged inter-administrative environments, where these applications are ex- pected to work, pose a series of problems (potentially wide number of partipants, low-reliability of nodes, absence of a centralized authority, etc.) that severely limit the scalability of existing approaches which were originally thought for supporting distributed applications built on the top of static and managed environments. In this paper we propose a novel architecture for implementing the topic-based publish/subscribe paradigm in large scale peer-to-peer systems. The proposed architecture is based on probabilistic mechanisms and peer-to-peer overlay man- agement protocols. It achieves event diffusion by implementing traffic confinement (published events have a high probability to reach only interested subscribers), high scalability (with respect to several fundamental parameters like number of participants, subscriptions, topics and event publication rate) and fair load distribution (load distribution closely follows the distribution of subscription on nodes). "

"Overlay management protocols have been introduced to guarantee overlay network connectivity in dynamic large-scale peer-to-peer systems. Some of these protocols have been specifically designed to avoid the partitioning of the overlay in large clusters (network breakage) despite massive node failures and the continuos arrivals/departures of nodes (churn). In this paper we identify a second effect connected to churn, namely network erosion. We show how erosion affects overlay network connectivity and point out that even a strongly connected overlay network, when exposed to continuous churn, can be disgregated in a relatively short time. More specifically the consequences of erosion are shown, through an experimental study, in the context of overlay management protocols based on view-exchange. We finally propose a connection recovery mechanism to be endowed at each node which is able to collaboratively detect node isolation and the presence of small clusters. This mechanism is shown to be effective in reducing the erosion of an overlay network exposed to continuous churn and to quickly recover its connectivity as soon as churn ends."

"Middleware for data distribution is a natural match and often a fundamental architectural building block for a large class of real-time, mission and safety critical application domains, such as industrial process control, air traffic control, defense systems, etc. This document contains an extended report about the activities done on the above said topics during the first year of collaboration between Universitá di Roma “La Sapienza” and Selex-SI. In this report we identify the main challenges in the realization of a scalable, QoS-driven, interoperable data distribution middleware. In particular, we identify three of the basic DDS features related respectively to scalable diffusion, timeliness and data availability. Then we analyze how their implementation could influence the interoperability protocol. "

"Point-to-point transmissions represent a fundamental primitive in any communication network. Despite many proposals have appeared in the literature, providing an ef- ficient implementation of such an abstraction in Mobile Ad Hoc Networks still remains an open issue. This paper proposes a probabilistic protocol for unicast packet delivery in a MANET. Unlike the classical routing protocols, in our proposal packet forwarding is not driven by a previously computed path. Rather, the nodes of the network exploit a set of routing meta-information (called hints) to discover a path to the destination on-the-fly. This assure robustness to topological changes, while requiring a very low overhead. A node gathers hints from the nodes located within a small number of hops (called the protocol’s lookahead) from itself. As showed through simulations, very good performance can be obtained with small lookahead. The main statistical properties of hints have been investigated through an analytical model, which is also reported in the paper."

"The advent of very large scale distributed applica tions in the form of peer-to-peer systems, introduced new problems caused by their intrinsic dynamic behav ior (i.e., nodes continuously joining and leaving the system). Such problems inhibit the usage of techniques developed for quasi-static distributed systems. In this paper we analyze the behavior of a simple subscription flooding based algorithm for publish/subscribe in a highly dynamic environment. Specifical ly we define the problem of keeping a list of processes matching a given event (Event Distribution List) complete and accurate. We propose a simple variant of the subscription flood ing approach, introducing expiration of subscriptions and their periodic refresh, and show how it can main tain accurate and complete EDLs in a dynamic dis tributed system."

"During the last decade, we assisted to an astounding revolution in the world. The widespread usage of internet-enabled applications, together with the advent of community-based interactions, completely changed our concept of collaborative work. One of the most important steps in this direction is the development of new technologies for data object storage, able to guarantee high degrees of reliability, while permitting the access in a nomadic environment through heterogeneous devices. In this chapter, we will study the problem of implementing a global data object storage system, exploring the current state of the art for correlated technologies, and surveying the most interesting proposals."

"During the last decade the publish/subscribe communication paradigm gained a central role in the design and development of a large class of applications ranging from stock exchange systems to news tickers, from air traffic control to defense systems. This success is mainly due to the capacity of publish/subscribe to completely decouple communication participants, thus allowing the development of applications that are more tolerant to communications asynchrony. This chapter introduces the publish/subscribe communication paradigm, stressing those characteristics that have a stronger impact on the quality of service provided to participants. The chapter also introduce the reader to two widely recognized industrial standards for publish/subscribe systems: the Java Message Service (JMS) and the Data Distribution Service (DDS)."

"An overlay network is formed on top of – and generally independently from – the underlying physical computer network, by the peers (nodes) of a P2P system. The dynamics of peers is taken into account by devising appropriate overlay maintenance protocols that are able to join and leave peers from the overlay. Due to the need for scaling in the number of nodes, overlay maintenance protocols have been simulated only in environments showing a very restricted behavior with respect to the possible concurrent and interleaved execution of join/leave operations. In this paper we compare two overlay maintenance protocols well suited to unstructured P2P systems, namely SCAMP and Cyclon, in an event-based simulation setting including concurrent and interleaved join and leave operations as well as variable message transfer delay. This simulation setting allows to point out surprising results for both protocols. In particular, under a continuous and concurrent replacement of nodes, permanent partitioning of the overlay arises after a very small number of join/leave operations."

"The decoupling and asynchrony properties of the content-based publish-subscribe paradigm makes it very appealing for dynamic wireless networks, like those that often occur in pervasive computing scenarios. Unfortunately, most of the currently available content-based publish-subscribe middleware do not fit the requirements of such extreme scenarios, in which the network is subject to very frequent topological reconfigurations due to the mobility of nodes. In this paper we propose a protocol for content-based message dissemination tailored to Mobile Ad Hoc Networks (MANETs) showing frequent topological changes. Message routing occurs without the support of any network-wide dispatching infrastructure thus eliminating the issue of maintaining such infrastructure on top of a physical network continuously changing its topology."

"Peer-to-peer systems (P2P) have become a popular technique to architect decentralized systems. However, despite its popularity most P2P systems consist in simple applications such as file sharing or chat systems. The main reason is that more complex applications require levels of consistency that nowadays are not offered by P2P systems. In this paper, we explore how to provide consistency based on distributed mutual exclusion via quorum systems in DHTbased P2P networks. Our results show that quorum systems applied directly to such networks are not scalable due to the high traffic imposed onto the underlying network. The paper introduces some design principles for both quorum systems and protocols using them that help to boost their performance. These design principles consist in dynamic and decentralized selection of quorums and in the exposition and exploitation of internals of the DHT such as the finger table. We show that by combining both design principles it is possible to minimize the number of visited sites and the latency needed to obtain a quorum."

"Point-to-point transmissions represent a fundamental primitive in any communication network. Despite many proposals have appeared in the literature, providing an efficient implementation of such an abstraction in Mobile Ad Hoc Networks still remains an open issue. This paper proposes a probabilistic protocol for unicast packet delivery in a MANET. Unlike the classical routing protocols, in our proposal packet forwarding is not driven by a previously computed path. Rather, the nodes of the network exploit a set of routing meta-information (called hints) to discover a path to the destination on-the-fly. This assure robustness to topological changes, while requiring a very low overhead. A node gathers hints from the nodes located within a small number of hops (called the protocol’s lookahead) from itself. As showed through simulations, very good performance can be obtained with small lookahead. The main statistical properties of hints have been investigated through an analytical model, which is also reported in the paper."

"The decoupling and asynchrony properties of the content-based publish-subscribe paradigm makes it very appealing for dynamic wireless networks, like those that often occur in pervasive computing scenarios. Unfortunately, none of the currently available content-based publish-subscribe middleware fit the requirements of such extreme scenarios in which the network is subject to very frequent topological reconfigurations due to the mobility of nodes. In this paper we propose a protocol for content-based message dissemination tailored to Mobile Ad Hoc Networks (MANETs) with frequent topological changes. Message routing occurs without the support of any network-wide dispatching infrastructure thus eliminating the issue of maintaining such logical topology on top of a time varying physical topology. The paper reports an extensive simulation study, which provides numerical evidence of the effectiveness of the approach."

"In this paper we present the results of a practical experience on the evaluation of two message-passing middleware platforms for developing distributed applications, i.e. the ACE/TAOReal Time Event Channel (RTEC) and the Maestro/Ensemble group communication toolkit (M/E). In particular, we compare their functionalities and their performances in a simple yet meaningful deployment configuration. The functional comparison points out the different characteristics of the two systems. In particular, M/E simplifies the coding of applications with strong requirements in terms of group membership tracking and ordered message delivery guarantees, while RTEC provides users with unreliable message delivery between loosely coupled processes. The performance comparison shows that, under stressing conditions, M/E sacrifices throughput stability for enforcing reliable and ordered message delivery, while RTEC offers a more stable throughput of unordered messages sacrificing message delivery reliability under heavy load. In normal operating conditions, the two systems perform almost similarly."

"Content-based routing in publish/subscribe systems is often realized through overlay networks with acyclic topologies. These types of network usually are simple to implement and maintain but are more prone to node crashes. In this research paper we introduce a Topology Management System (TMS), a component of a content-based pub/sub broker. Aim of the TMS is to mask dynamic changes of the underlying topology to the content-based routing engine. TMS is resilient to certain broker failures and allows brokers join and leave."

"Content-based routing realized through static networks of brokers usually exhibit nice performance when subscribers with similar interest are physically close to each other (e.g. in the same LAN, domain or in the same geographical area) and connected to a broker which is also nearby. If these subscribers are dispersed on the Internet, benefits of such routing strategy significantly decrease. In this paper we present a Topology Management System (TMS) which is a component of a content-based pub/sub system. The aim of a TMS is to mask dynamic changes of the brokers' topology to the content-based routing. TMS relies on a self-organizing algorithm whose goal is to place close (in terms of TCP hops) brokers that manage subscribers with similar interests keeping acyclic the topology and without compromising network-level performance. TMS is also resilient to broker failures and allows joining of new brokers."

"In this paper we outline a novel approach for enhancing content-based routing performance in publish/subscribe systems composed by an application-level network of event brokers. We present an algorithm which aim is to logically place close (in terms of TCP hops) brokers that manage subscribers with similar interests. We show how this algorithm allows the content based routing algorithm to work in a favorable scenario, thus increasing its global performance."