Overlay Network

Although the number of accidents that occur on roads is slowly decreasing in time, as both cars and streets become progressively safer, the Statistics of Road Traffic Accidents study from the United Nations reveals that during the last decade every year an average of 150,000 people have lost their lives and 5.5 million have suffered injuries on the roads of western countries. Among the many proposals that have been made, during the years, to combat such phenomenon, an important place is taken by highway accident warning technologies. Such type of system can play an important role, especially in those cities, like Los Angeles, Seoul and Shanghai, specifically designed for cars, more than for human beings. In fact, when an accident occurs on a highway of a city like Los Angeles, it is vital to warn as rapidly as possible all the approaching vehicles, in order to give them the time to stop before crashing on any unexpected obstacle. Recently, an inter-vehicular accident warning system has been devised, theoretically proven to be optimal in terms of bandwidth usage and covered distance. In this paper, we present preliminary results that assess the feasibility of such system. The presented results and measurements were taken from real experiments, performed on Los Angeles freeways and roads in August 2011. To the best of our knowledge, these are the first real experiments of such magnitude performed in a live setting with an inter-vehicular accident warning system.

Continuous monitoring of a network domain poses several challenges. First, routers of a network domain need to be polled periodically to collect statistics about delay, loss, and bandwidth. Second, this huge amount of data has to be mined to obtain useful monitoring information. This increases the overhead for high speed core routers, and restricts the monitoring process from scaling to a large number of flows. To achieve scalability, polling and measurements that involve core routers should be avoided. We design and evaluate a distributed monitoring scheme that uses only edge-to-edge measurements, and scales well to large network domains. In our scheme, all edge routers form an overlay network with their neighboring edge routers. The network is probed intelligently from nodes in the overlay to detect congestion in both directions of a link. The proposed scheme requires significantly fewer number of probes than existing monitoring schemes. Through analytic study and a series of experiments, we show that the proposed scheme can effectively identify the congested links. The congested links are used to capture the misbehaving flows that are violating their service level agreements, or attacking the domain by injecting excessive traffic.

MotusNet is a content network in which the addressable objects, contents, are decoupled from the physical nodes that carry them. The content names in MotusNet are hierarchical, so that they can be aggregated in the routing tables to support a larger ...

Kubernetes is the leading platform for orchestrating containerized applications. In this paper, we extend Kubernetes networking to make use of SRv6, a feature-rich overlay networking mechanism. Integration with SRv6 can be very beneficial when Kubernetes is used in large-scale and distributed multi-datacenter scenarios. We have focused on the Calico CNI plugin, one of the most used Kubernetes networking plugins. In particular, we consider Calico-VPP, a version of the Calico plugin based on the VPP (Vector Packet Processing) data plane, which provides support for SRv6 operations with very high performance. The proposed SRv6 overlay networking solution for Kubernetes offers several advantages compared to a traditional overlay (e.g. IP in IP), in particular the possibility to use Traffic Engineering for the overlay tunnels. In the paper, we provide the architecture and the detailed design of the SRv6 based overlay and describe our open source implementation. We consider the research and technological question on how to extend Kubernetes networking to support large-scale and distributed multi-datacenter scenarios, which is an important goal for Cloud and Network providers. In this respect, we compare two different solutions for the control plane architecture of the SRv6 capable Kubernetes networking plugin, one based on the BGP routing protocol and another one based on extending the Kubernetes control plane. Finally, we report a performance evaluation of the data plane of the proposed SRv6 overlay networking, showing that it has comparable performance to existing overlay solutions (e.g. IP in IP), while offering a richer set of features.

Computational grids are believed to be an effective and scalable solution to the problem of resource sharing over large, heterogeneous networks of computing devices. Since grids are highly distributed in nature, one of the most challenging problems is the discovery of dynamic resources in a grid. In this paper we use ideas from P2P systems to propose a solution for the problem. Specifically, we classify nodes as consumers and producers, depending on whether they consume or produce more jobs. Our algorithm connects all producer nodes using a overlay network that is a small-world graph (the graph is produced by adding "shortcut" chords to a circle). The consumer nodes hang off the small world graph. The producer nodes are forced to take part in resource cataloging and discovery. This has three distinct advantages -first, it prevents "freeloading" by forcing producers to do useful work; second, it frees the consumers to only do computations; third, the low diameter of the overlay graph ensures that all resources are within a small number of hops. We simulate and evaluate the performance of our algorithm in realistic traffic conditions. We evaluate the performance of our algorithm using metrics like the average time to answer the query, the average number of requests that were dropped and the average number of hops traveled by query packets. Our experiments show that our algorithm performs well with thousands of nodes.

Session Initiation Protocol (SIP) is a control-plane protocol for multiple services such as VoIP, Instant Messaging and Presence, and in addition, is key to IP Multimedia Subsystem (IMS). A SIP message consists of plain-text headers and their corresponding values, which are used to route the message between one or more endpoints, resulting in a media session. These headers and values are often transformed/re-written at intermediate SIP servers ("proxies"). It is important to monitor the flow and transformation of such messages in real-time, for functional testing of a SIP overlay network containing malfunctioning or ill-configured SIP entities, or for efficient run-time SIP network operation, including problem determination and load balancing. Towards that end, we have designed and implemented a programmable in-kernel Linux SIP message classification engine. The classifier can be configured to intercept incoming and outgoing SIP messages from a server, extract appropriate message meta-data including distinguishing header-value pairs and their transformations, and forward the same to a monitoring engine. The engine collates this information from different classifiers across the network, to infer the state of a SIP call on individual servers on the call path as well as aggregated call-state.

Distributed knowledge management systems (DKMS) have been suggested to meet the requirements of today's knowledge management. Peer-to-peer systems offer technical foundations for such distributed systems. To estimate the value of P2P-based knowledge management evaluation criteria that measure the performance of such DKMS are required. We suggest a concise framework for evaluation of such systems within different usage scenarios. Our approach is based on standard measures from the information retrieval and the databases community. These measures serve as input to a general evaluation function which is used to measure the efficiency of P2P-based KM systems. We describe test scenarios as well as the simulation software and data sets that can be used for that purpose.

Almost all the natural or human made systems can be understood and controlled using complex networks. This is a difficult problem due to the very large number of elements in such networks, on the order of billions and higher, which makes it impossible to use conventional network analysis methods. Herein, we employ artificial intelligence (specifically swarm computing), to compute centrality metrics in a completely decentralized fashion. More exactly, we show that by overlaying a homogeneous artificial system (inspired by swarm intelligence) over a complex network (which is a heterogeneous system), and playing a game in the fused system, the changes in the homogeneous system will reflect perfectly the complex network properties. Our method, dubbed Game of Thieves (GOT), computes the importance of all network elements (both nodes and edges) in polylogarithmic time with respect to the total number of nodes. Contrary, the state-of-the-art methods need at least a quadratic time. Moreover, the excellent capabilities of our proposed approach, it terms of speed, accuracy, and functionality, open the path for better ways of understanding and controlling complex networks. In any real-world system, at micro and macro-scale, from the vigintillions of interacting atoms in the observable universe, to the billions of persons who live on Earth, there are amazing structures of networks of networks. These networks can be studied, understood, and controlled by the means of network science and complex networks 1 , leading to advances in many domains, including neuroscience 2-4 , astrophysics 5 , biology 6,7 epidemiology 8 , social networks 9,10 , transportation networks 11 , communication networks , and artificial intelligence 14 (to mention but a few). Yet, unveiling the complex networks hidden patterns and computing even their most basic properties is far from trivial, due to the massive number of node entangles that interact in non-obvious ways, evolving and unfolding continuously 15 . Among all these network properties, the centrality (or importance) of nodes and links is fundamental to understanding things such as: biological neural networks 2-4 , cosmic structures 5 , biological networks 7 , how viruses spread or can be contained 16 ; which people or news are influencing opinions and decisions the most 17 ; how to protect computer systems from cyber-attacks 18 ; or how to relay data packets in the one-trillion Internet-of-Things network of the future. While there is ample literature on node centrality computation 19 , the existing methods do not scale to the size and dynamics of practical complex networks, which operate at the tunes of millions to trillions nodes. Besides that, the state-of-the-art centrality metrics are designed for specific goals, and one metric which performs well for one goal is suboptimal for another 20 . Furthermore, existing methods focus on finding the most important network elements (i.e. nodes or links), but fail to capture the hidden relations across the whole network links and nodes. The centralized algorithms consider the topology as a whole, overlooking many of the local features . Per contra, the decentralized methods are usually based on local computations to construct statistics of network elements (as in 21 ), but fail to capture the overall network structure. In fact, the most effective decentralized methods nowadays still fail to capture all the relations between the networks elements, and this is our main target. In addition, current methods have technological constraints that have to be surpassed. To tackle the scale as well as dynamics of real-world networks, we need to compute centrality metrics not only accurately but also timely, based on the existing computational capabilities.

Almost all the natural or human made systems can be understood and controlled using complex networks. This is a difficult problem due to the very large number of elements in such networks, on the order of billions and higher, which makes it impossible to use conventional network analysis methods. Herein, we employ artificial intelligence (specifically swarm computing), to compute centrality metrics in a completely decentralized fashion. More exactly, we show that by overlaying a homogeneous artificial system (inspired by swarm intelligence) over a complex network (which is a heterogeneous system), and playing a game in the fused system, the changes in the homogeneous system will reflect perfectly the complex network properties. Our method, dubbed Game of Thieves (GOT), computes the importance of all network elements (both nodes and edges) in polylogarithmic time with respect to the total number of nodes. Contrary, the state-of-the-art methods need at least a quadratic time. Moreover...

This paper presents an analysis on performance of logarithmic degree structured P2P (peer-to-peer) overlay networks. P2P network consist of highly transient peers, where peers join and leave the network randomly also known as dynamic environment. It is, therefore difficult to measure the performance of the parameters in real environment. The design of structured overlay networks is fragmented and due to various designs few simulations have been conducted to compare the protocols in dynamic environment. The outcome of the analysis helps the decision in choosing and designing better structured overlay protocol for P2P network. In order to evaluate the routing performance, this work simulates logarithmic-hop overlays -Chord, Pastry and Kademlia. The result shows that among Chord, Pastry and Kademlia protocols, performance of Kademlia is better than Chord and Pastry with 94.2 -99% routing efficiency. Hence, Kademlia architecture is better choice to implement structured P2P network.

Overlay routing is known to cause undesired instability in a network by operating in a selfish manner. The objectives of overlay routing, such as optimizing end-to-end latency, are often in conflict with the objectives of traffic engineering in the native layer, which is concerned about balancing load. In our work, we build on past research that has investigated the recurring non-cooperative interaction between overlay routing and traffic engineering, and develop strategies that improve the routing performance of a particular layer with incomplete information about the other layer. In our strategies, one layer acts as a leader that predicts the follower's reaction and undertakes countermeasures to prevent future deterioration in performance. Specifically, we propose two classes of strategiesfriendly or hostile -for each layer. By simulating under different network characteristics, we show that these preemptive strategies achieve near-optimal performance for the leader and increase the overall stability of the network. Furthermore, we observe that the best performance for a particular layer is achieved only when the goals of the other layer are completely violated, thereby motivating a higher level of selfishness.

The onion routing (Tor) protects the IP and behavioral anonymity on the Internet by routing the networking packet through multiple randomly selected relay nodes between the source and the destination. It therefore adds transmission delays, and the latency cost is greater than not using Tor (i.e., when the packet is transmitted directly to the destination). Machine learning can be used to learn about the Tor circuit delays, given the source and destination, to identify the circuits with smaller latency cost. In this paper, however, we show how using reinforcement learning in its by-default, deterministic fashion can deprive anonymity. More specifically, choosing the performancebest circuit corresponding to the exploitation strategy provides no anonymity with an entropy of zero. Such approach can eliminate the anonymity protection and performs worse than disabling Tor altogether. We therefore recommend a strategic and appropriate application of machine learning which not only advances the performance but also take the anonymity into account for future research to advance Tor and Internet anonymity.

The diffusion of wireless technologies is rapidly growing, and they are increasingly employed to provide connectivity on the road. Wireless systems are widely deployed to allow cheap devices to communicate, to provide local networking capabilities to even more demanding customers, e.g. for short range on-line gaming. One of the key problems is related to provide some Quality of Service (QoS) mechanisms in order to partition and control the underlying physical resource. Even if mechanisms for supporting QoS are becoming available also at the Medium Access Control (MAC) layer, this paper introduces an approach independent of the MAC, called PRIMUS (PRoviding QoS Independently of MAC UserS). The proposed scheme is based on a peer-to-peer (p2p) framework and it exploits an overlay network model in order to be fully decoupled from the underlying physical technology.

Recent distributed systems, in a P2P scenario, are characterized by a set of independent peers that dynamically need to cooperate by sharing data and services. For effective collaboration, under highly dynamic conditions and in absence of a global view of the resources shared across the information systems, semantic modeling tools need be defined. In particular, the emergence of semantics is a key issue to enforce timely discovery and integration of distributed data and services. In the paper we discuss, in particular, the construction of a semantic overlay for service sharing and discovery in service-oriented applications in a P2P scenario. The semantic overlay is built over the P2P network: each peer has ontology-based service descriptions; semantic links between similar service descriptions belonging to different peers are established and maintained according to the network evolution; the result is P2Pintegrated knowledge space (here considered for services but extendible to data). The semantic overlay can be seen as a continuously evolving conceptual map across collaborative peers that provide similar services and constitute synergic service centres in a given domain. The semantic links enable effective similarity-based service search and optimization strategies are defined for request propagation over the P2P network. From the conceptual modeling perspective, the semantic overlay can be considered as an evolved conceptual representation where specific modeling requirements due to new technologies, service-oriented technology and P2P technology, are considered. The Evolution of Conceptual Modeling

We present YAAP: a generic active network architecture with the ability to manually and dynamically deploy network services within distributed network nodes and providing a secure execution environment (EE). It also demultiplexes active network packets to multiple EEs located on the same network node. Currently, YAAP prototype is implemented under the Linux operating system, with some parts built in kernel-space and others in user-space to optimize performance. This paper describes the node architecture, serves as an internal description of the implementation, and also demonstrates the utility of YAAP's activeness by experimentally deploying active applications using the current implementation.

This paper presents a contribution to the OMNeT++ 4.1 simulator in C++ by integrating our own implementation of the FF -Ford-Fulkerson multipath routing algorithm. In order to validate it, an eight-node testbed works with dynamic requested and released flows on each link, updated every second. Four simple scenarios were proposed, considering all possible combinations of fixed and/or random variation laws. Additionally, the path search algorithm BFS -Breadth First Search was implemented too. This software package running under Fedora Core gets the input data from three files called Nodes, Connections and Params. This mechanism provides the interworking between the C++-based FF simulator and the Java-based application in Intellij IDEA 10.5, used as traffic variation simulator. The graphical representations of the occupied, the requested and, respectively, the released flows are obtained by aggregating the contributions of all flows from the paths for any given source-destination pair.

This paper addresses the problem of capacity pro- visioning in a QoS enabled IP network, subject to satisfying varying performance requirements for different traffic classes. These performance requirements are viewed in terms of mean end-to-end delays required for the various traffic classes or in terms of random variations of their delays (i.e., jitter) or a combination of both. The accuracy of the model is investigated by means of a simulation study over an extended range of test cases for different traffic load characteristics and traffic intensities at the nodes. The results demonstrate the capability of the model in guaranteeing the end-to-end delay requirements for the traffic classes. procedures that allow burstiness of multi-class IP traffic to be effectively modelled. The selection of a renewal traffic model i.e., GI arrival process, for this purpose, represents a reasonable bal- ance between the accuracy (modelling power) and the efficiency required by a network planning t...

Multicasting is the enabling technology for group communication. However, network-layer multicasting (e.g., IP multicast) has not been widely adopted more than 10 years of its invention due to the concerns related to deployment, scalability and network management. Application-layer multicast (ALM) has been proposed as an alternative for IP multicast. In ALM, group communications take place on an overlay network in which each edge corresponds to a direct unicast path between two group members. ALM protocols differ in, among other aspects, the topology of the underlying overlay network (e.g., tree, mesh or ring). Ring-based ALM protocols have the advantages of providing a constant node degree, and enabling the implementation of reliable and totally-ordered message delivery through the use of a ring with a token that contains ordering and flow control information. In addition, a ring overlay network topology is inherently reliable to single node failures. In this paper, we provide a survey and a taxonomy of several ring-building group communication protocols. Investigating the major characteristics of ring-building network protocols is an important step towards understanding which of them are suitable for command and control group communications.

This paper presents the design, analysis and performance evaluation of VRing, a novel application-layer multicast (ALM) protocol that establishes a virtual ring as an overlay network among the multicast group members in a selforganizing and distributed manner. VRing takes advantage of the inherent desirable property of a ring; namely, the degree of each node on the ring is O´1µ (i.e., independent of the number of the group members) and hence, the state maintained at each group member is also O´1µ. In order to reduce the routing delay of the ring overlay network, we propose to form a spare ring overlay structure that improves connectivity among group members. The design of the spare ring is justified by proposing, and analytically studying the performance of, a data delivery and duplicate suppression mechanism that makes use of both the original ring and the spare ring for forwarding data packets. We conduct simulations of both VRing and a hierarchical ALM protocol, NICE, using the J-Sim network simulator. Simulation results show that although VRing has a higher path stretch and a higher link stress than NICE, it provides less control overhead, consumes less bandwidth, and provides lower average node degree than NICE. Furthermore, VRing achieves a higher average data delivery ratio in the presence of failures than NICE. The performance improvement is especially pronounced for larger multicast groups.

Application-level peer-to-peer (P2P) network overlays are an emerging paradigm that facilitates decentralization and flexibility in the scalable deployment of applications such as group communication, content delivery, and data sharing. However the construction of the overlay graph topology optimized for low latency, low link and node stress and lookup performance is still an open problem. We present a design of an

Real-world IP applications such as Peer-to-Peer filesharing are now able to benefit from network and location awareness. It is therefore crucial to understand the relation between underlay and overlay networks and to characterize the behavior of real users with regard to the Internet. For this purpose, we have designed and implemented MULTI-PROBE, a framework for large-scale P2P file-sharing measurements. Using this framework, we have performed measurements of BitTorrent, which is currently the P2P file sharing network with the largest amount of Internet traffic. We analyze and correlate these measurements to provide new insights into the topology, the connectivity, and the path characteristics of the Internet parts underlying P2P networks, as well as to present unique information on the BitTorrent throughput and connectivity.

Real-world IP applications such as Peer-to-Peer filesharing are now able to benefit from network and location awareness. It is therefore crucial to understand the relation between underlay and overlay networks and to characterize the behavior of real users with regard to the Internet. For this purpose, we have designed and implemented MULTIPROBE, a framework for large-scale P2P file-sharing measurements. Using this framework, we have performed measurements of BitTorrent, which is currently the P2P file sharing network with the largest amount of Internet traffic. We analyze and correlate these measurements to provide new insights into the topology, the connectivity, and the path characteristics of the Internet parts underlying P2P networks, as well as to present unique information on the BitTorrent throughput and connectivity.

The growth of the Internet and its popular services are forcing telecom operators to provide advanced services to their subscribers, as traditional voice services are no longer enough to attract more customers. To enable more innovative and value-added IP services and take advantage of the services that the Internet provides, the IP Multimedia Subsystem(IMS) is introduced. The IMS provides a complete access-agnostic architecture and framework that facilitates the convergence of the mobile network, removing the gap between the two most successful communication networks that are the cellular and the Internet network. The harmonized All-IP platform has the potential to provide all Internet services with a more cost-effective and more efficient architecture than the circuit-switched networks do. However, by merging two of the most successful networks, the integration of two network models with different concerns and motivations is not without its problems, among which the scalability issue is the most essential, when supporting content delivery services. The purpose of this chapter is to study and design a new content delivery network infrastructure, PeerMob, merging the Peer-to-Peer technology with the IMS framework, which benefits IMS with scalability, reliability, and efficiency features coming with decentralized P2P architecture. The chapter also puts this P2P IMS paradigm under realistic network conditions and strenuous simulation to evaluate the performance of the P2P IMS system.

Traditional software and security patch update delivery mechanisms rely on a client/server approach where clients pull updates from servers regularly. This approach, however, suffers a high window of vulnerability (WOV) for clients and the risk of a single point of failure. Overlay-based information dissemination schemes overcome these problems, but often incur high infrastructure cost to set up and maintain individual information dissemination networks. Against

Both Mobile Ad Hoc Networks (MANETs) and Peer-to-Peer (P2P) Overlay Networks are decentralized and self-organizing networks with dynamic topology and responsible for routing queries in a distributed environment. Nevertheless, MANETs are composed of resource-constrained devices susceptible to faults, whereas P2P networks are fault-tolerant. Thus, we argue that these systems are not only similar, but also complementary, and that P2P networks are the ideal data sharing system for MANETs. This paper tackles the problem of integrating both networks from the point of view of content discovery. For that, we confront two classes of strategies: the structured strategy -based on distributed indexing -with the unstructured -based on flooding searches. More specifically, we study the performance of two P2P content discovery protocols over a MANET: Chord (structured) and Gnutella (unstructured). We use simulation to evaluate both protocols under many aspects and conclude that the choice of the strategy to be used depends on the application characteristics and the environment where it would be employed. Resumo. Ambas as Redes Móveis Ad hoc (MANETs) e as Redes Par-a-Par (P2P) são auto-organizáveis, de natureza descentralizada e responsáveis por rotear requisic ¸ões em ambientes distribuídos de topologia dinâmica. Entretanto, enquanto as MANETs são compostas de dispositivos com poucos recursos e, portanto, suscetíveis a falhas, as redes P2P são conhecidas pela sua tolerância a falhas. Por esta razão, sustentamos que MANETs e redes P2P não são apenas similares, mas também complementares e que sistemas P2P são ideais para se compartilhar arquivos em MANETs. Este trabalho aborda a integrac ¸ão dessas duas redes do ponto de vista de localizac ¸ão de conteúdo. Para tal, confrontamos a estratégia estruturada -que utiliza índexac ¸ão distribuídacom a não estruturada -que emprega flooding nas buscas. Mais especificamente, estudamos o desempenho sobre uma MANET de dois protocolos P2P de localizac ¸ão de conteúdo: Chord (estruturado) e Gnutella (não-estruturado). Utilizamos simulac ¸ão para efetuar os estudos sob diversos cenários e os resultados indicam que a escolha da estratégia a empregar dependerá da aplicac ¸ão e do ambiente em que será utilizada. O advento da computac ¸ão móvel e o avanc ¸o da Internet nos últimos anos possibilitou o aparecimento de novas tecnologias e a quebra de paradigmas. Dentre tais tecnologias duas têm obtido especial atenc ¸ão devido à vasta gama de aplicac ¸ões que oferecem e à facilidade de * Trabalho desenvolvido com apoio do CNPq/MCT e da CAPES/MEC.

We describe a new scalable application-layer multicast protocol, specifically designed for low-bandwidth, data streaming applications with large receiver sets. Our scheme is based upon a hierarchical clustering of the application-layer multicast peers and can support a number of different data delivery trees with desirable properties.We present extensive simulations of both our protocol and the Narada application-layer multicast protocol over Internet-like topologies. Our results show that for groups of size 32 or more, our protocol has lower link stress (by about 25%), improved or similar end-to-end latencies and similar failure recovery properties. More importantly, it is able to achieve these results by using orders of magnitude lower control traffic.Finally, we present results from our wide-area testbed in which we experimented with 32-100 member groups distributed over 8 different sites. In our experiments, average group members established and maintained low-latency paths and ...

Researchers are directing enormous efforts to achieve the aim of global mobility by enhancing the standard mobile IP with various routing schemes focusing on best routes and least cost while ignoring the facts of the organizations usual use of private IP’s and the presence of firewalls. Nevertheless, existing Mobile IP models are still missing three basic concepts that hinder their applicability in real environment. First, global mobility must be independent of the different infrastructure technologies (e.g., Wi-Fi, WiMAX, UMTS, etc.). Second, a secure authentication mechanism for guiding the access of mobile nodes to the corporate network’s resources is certainly needed. Third, the capability of correlating the mobile node’s activities to a real world identity is a requirement of security in a wider since i.e., network, web, and national security. This paper defines the role of global mobility in facilitating and improving mobile business performance. It, also, presents a review of...

Despite its popularity, relatively little is known about the traffic characteristics of the Skype VoIP system and how they differ from other P2P systems. We describe an experimental study of Skype VoIP traffic conducted over a five month period, where over 82 mil-lion datapoints were ...

They appeared in our life some years ago with the awakening of the PC and now the are everywhere : computers have become ubiquitous and, almost, irreplaceable. Classical ways of creating, managing and exchanging information have been progressively replaced by electronic means. In spite of this plebiscite, computer supported collaborative work (CSCW) softwares can be blamed for requiring the user to do an effort to use them. This paper describes an artificial ant based autonomous information dissemination algorithm. It constitutes the communication layer of our framework PIAF ("Personal Intelligent Agent Framework") intended to help users transparently sharing information. The algorithm uses message gossiping strategy to transfer information items between users.

Adequately supporting applications with Quality-of-service (QoS) requirements is a challenging task, particularly in the Internet. In great part, this is due to the fact that the provisioning of end-toend QoS to traffic that traverses multiple autonomous systems (ASes) requires a level of cooperation between ASes that is difficult to achieve in the current architecture. Recently, service overlay networks have been considered as an approach to QoS deployment that avoids these difficulties. In this study, the problem of the topological synthesis of a service overlay network is addressed. Endsystems and nodes of the overlay network are connected through ISPs that supports bandwidth reservations. The topology design problem is expressed as an optimization problem. Even though the design problem is related to the (in general NP-hard) quadratic assignment problem, it is shown that relatively simple heuristic algorithms can deliver results that are sometimes close to the optimal solution with little overhead.

In this work, we have considered a new hierarchical non-DHT based architecture for Peer-to-Peer (P2P) networks in which at each level of the hierarchy existing networks are all structured and each such network has the diameter of 1 overlay hop. Such low diameters have immense importance in designing very efficient data lookup algorithms. The mathematical model based on the Chinese Remainder Theorem has been used to define the neighborhood relations among peers to obtain the above-mentioned diameters. In this work, we present very efficient intra-group as well as inter-group data look up algorithms with O(1) time complexity exploiting the above mentioned low diameter features of the P2P system. Besides, we present very efficient algorithms related to peers joining with new and existing resource types.

The present paper proposes an architecture for a pure peer-to-peer communication system that is free from centralized coordination and knowledge. To meet this decentralization requirement the current paper focuses on a novel Overlay Discovery Algorithm which allows peers to connect to the network in an efficient and dynamic fashion. The system consists of three core modules which enable peers to meet, organize and communicate respectively. In addition, a series of simulation results is presented as a proof of concept for the Overlay Discovery Algorithm.

This paper presents a fast near linear run-time algorithm for deadline-satisfying packet forwarding and route lookup. This algorithm has been designed after a Constraint Resource Planning (CRP) methodology of approximation algorithm design. The algorithm uses scheduling and forwarding heuristics to produce near- optimal performance. These heuristics execute within a generic shell. In this paper we present the algorithm, the shell

Although Peer-to-Peer live video streaming over wireless mesh networks (WMNs) is considered a promising technology, some important challenges such as interference, mobility and limited available resources in gadgets (e.g. Smartphones and Tablets) may significantly reduce the perceived video quality. GREENIE and MATIN, in our previous studies, provided an efficient routing protocol in WMNs and a video streaming method based on random network coding (RNC), respectively. Therefore, their integration in the form of an enhanced framework, named GAZELLE, can considerably increase the video quality on these gadgets by decreasing the video distortion, dependency distortion, initial start-up delay and end-to-end delay. Findings using a precise simulation in OMNET?? show that GAZELLE noticeably outperforms other frameworks. GAZELLE not only decreases the imposed computational complexity and transmission overhead due to using RNC considerably, but it also efficiently routes video packets through those gadgets which does not require neither high battery energy sources nor high CPU power.

In recent years, Random Network Coding (RNC) has emerged as a promising solution for efficient Peer-to-Peer (P2P) video multicasting over the Internet. This probably refers to this fact that RNC noticeably increases the error resiliency and throughput of the network. However, high transmission overhead arising from sending large coefficients vector as header has been the most important challenge of the RNC. Moreover, due to employing the Gauss-Jordan elimination method, considerable computational complexity can be imposed on peers in decoding the encoded blocks and checking linear dependency among the coefficients vectors. In order to address these challenges, this study introduces MATIN which is a random network coding based framework for efficient P2P video streaming. The MATIN includes a novel coefficients matrix generation method so that there is no linear dependency in the generated coefficients matrix. Using the proposed framework, each peer encapsulates one instead of n coefficients entries into the generated encoded packet which results in very low transmission overhead. It is also possible to obtain the inverted coefficients matrix using a bit number of simple arithmetic operations. In this regard, peers sustain very low computational complexities. As a result, the MATIN permits random network coding to be more efficient in P2P video streaming systems. The results obtained from simulation using OMNET++ show that it substantially outperforms the RNC which uses the Gauss-Jordan elimination method by providing better video quality on peers in terms of the four important performance metrics including video distortion, dependency distortion, End-to-End delay and Initial Startup delay.

The scarcity of the radio channel is the main bottleneck toward maintaining the quality of service (QoS) in a mobile cellular network. As channel allocation schemes become more complex and computationally demanding, alternative computational models that include knowledge-based algorithms and provide the means for faster processing are becoming a topic of research interest. An efficient deterministic technique, capable of handling channel allocation problems, is introduced as an alternative. The proposed model utilizes the Global Positioning System (GPS) data for tracing the hosts’ likely movements within and across the cells and allocates the channels to the mobile devices accordingly. The allocation of the channels to the mobile hosts is deterministic in the sense that the decision of the channel allocation is based on the realistic data received from the GPS about the hosts’ movements. The performance of the proposed technique has been evaluated by conducting the simulation experi...

Computation in the mobile grid nodes under the cellular network environment requires an efficient management of wireless channels along with the user mobility. Due to random movement of the mobile devices, the load over the cells in terms of initiation of new computation or carryover computation (handoff) may vary dynamically. This may result in non-availability of free channels for Inter-task communication leading in the drop of carryover computation or to initiate the new computation in the current cell. The proposed work designs a model by instigating substantive cooperation among underutilized and the overloaded cells, considering importance to the frequency reuse and assigning priority to the on-going computation in the computational mobile grid. The model seeks cooperation by grouping the cells in different sizes to reduce the blocking and dropping of the computation. Blocking of the communication is very serious in computational mobile grid environment as the drop may result ...

The existing peer-to-peer networks have several problems such as fake content distribution, free riding, white-washing and poor search scalability, lack of a robust trust model and absence of user privacy protection mechanism. Although, several trust management and semantic community-based mechanisms for combating free riding and distribution of malicious contents have been proposed by some researchers, most of these schemes lack scalability due to their high computational, communication and storage overhead. This paper presents a robust trust management scheme for P2P networks that utilizes topology adaptation by constructing an overlay of trusted peers where the neighbors are selected based on their trust ratings and content similarities. While increasing the search efficiency by intelligently exploiting the formation of semantic community structures by topology adaptation among the trustworthy peers, the scheme provides the users a very high level of privacy protection of their usage and consumption patterns of network resources. Simulation results demonstrate that the proposed scheme provides efficient searching to good peers while penalizing the malicious peers by increasing their search times as the network topology stabilizes.

One major benefit of named-data networking (NDN) is its potential to control network load by leveraging in-network caching and request aggregation. Both the network operator and consumers benefit from these features, as operating costs are reduced and quality-of-experience is increased. However, request aggregation, combined with NDN's loop prevention mechanisms, can create denial-of-service (DoS) against client interests (intentionally and unintentionally) by clients employing multicast forwarding. In this paper, we discuss this problem and propose three increasingly efficient solutions to address the problem; our arguments are backed by simulation and numerical analyses.

creates new opportunities to save costs by converging data and telephone services. The primary question of our study is whether emerging metropolitan networks can meet the QoS requirements necessary to connect GSM and UMTS base stations. These requirements on delay, jitter, and loss are significantly more stringent than the ones for VoIP and have to be met at the presence of bursty cross traffic. Therefore, we have probed ETH's campus network which spans the metropolitan area of Zurich with traffic comparable to encapsulated E1 traffic from base stations and have measured perceived QoS. Our findings show that the campus network generally meets the QoS requirements. However, the perceived QoS degrades with increasing network utilization. To further investigate the impact of the configuration of Ethernet switches on perceived QoS, we have conducted a simulation study. The results show that the perceived QoS is better when switch buffers are limited to small sizes of 1 MB compared to setups where the number of frames in the buffer is limited. From these results we infer that metropolitan Gigabit Ethernets are well suited for connecting GSM and UMTS base stations.

Path probing is essential to maintaining an efficient overlay network topology. However, the cost of complete probing can be as high as ¢ ¤£ ¦¥ ¨ § © , which is prohibitive in large- scale overlay networks. Recently we proposed a method that trades probing overhead for inference accuracy in sparse networks such as the Internet. The method uses physical path information to infer path quality for all of the ¥ £ ¦¥ © overlay paths, while actually probing only a subset of the paths. In this paper we propose and evaluate a distributed approach to implementing this method. We describe a minimum diameter, link-stress bounded overlay spanning tree, which is used to collect and disseminate path quality information. All nodes in the tree collaborate to infer the quality of all paths. Simulation results show this approach can achieve a high-level of inference accuracy while reducing probing overhead and balancing link stress on the spanning tree.

This paper describes Service Clouds, a distributed infrastructure designed to facilitate rapid prototyping and deployment of services that enhance communication performance, robustness, and security. The infrastructure combines adaptive middleware functionality with an overlay network substrate in order to support dynamic instantiation and reconfiguration of services. The Service Clouds architecture includes a collection of lowlevel facilities that can be either invoked directly by applications or used to compose more complex services. After describing the Service Clouds architecture, we present results of two experimental case studies conducted on the PlanetLab Internet testbed, the first to improve throughput of bulk data transfer, and the second to enhance the robustness of multimedia streaming.

In pervasive computing environments, conditions are highly variable and resources are limited. In order to meet the needs of applications, systems must adapt dynamically to changing situations. Since adaptation at one system layer may be insufficient, crosslayer, or vertical approaches to adaptation may be needed. Moreover, adaptation in distributed systems often requires horizontal cooperation among hosts. This cooperation is not restricted to the source and destination(s) of a data stream, but might also include intermediate hosts in an overlay network or mobile ad hoc network. We refer to this combined capability as universal adaptation. We contend that the model defining interaction between adaptive middleware and the operating system is critical to realizing universal adaptation. We explore this hypothesis by evaluating the Kernel-Middleware eXchange (KMX), a specific model for crosslayer, cross-system adaptation. We present the KMX architecture and discuss its potential role in supporting universal adaptation in pervasive computing environments. We then describe a prototype implementation of KMX and show results of an experimental case study in which KMX is used to improve the quality of video streaming to mobile nodes in a hybrid wired-wireless network.

Use of multiple paths between node pairs can enable an overlay network to bypass Internet link failures. Selecting high quality primary and backup paths is challenging, however. To maximize communication reliability, an overlay multipath routing protocol must account for both the failure probability of a single path and link sharing among multiple paths. We propose a practical solution that exploits physical topology information and end-to-end path quality measurement results to select high quality path pairs. Simulation results show the proposed approach is effective in achieving higher multipath reliability in overlay networks at reasonable communication cost.

Path probing is essential to maintaining an efficient overlay network topology. However, the cost of a full-scale probing is as high as ¢ ¤£ ¦¥ ¨ § © , which is prohibitive in large-scale overlay networks. Several methods have been proposed to reduce probing overhead, although at a cost in terms of probing completeness. In this paper, an orthogonal solution is proposed that trades probing overhead for estimation accuracy in sparse networks such as the Internet. The proposed solution uses network-level path composition information (for example, as provided by a topology server) to infer path quality without full-scale probing. The inference metrics include latency, loss rate and available bandwidth. This approach is used to design several probing algorithms, which are evaluated through analysis and simulation. The results show that the proposed method can significantly reduce probing overhead while providing bounded quality estimations for all ¥ £ ¦¥ © overlay paths. The solution is well suited to medium-scale overlay networks in the Internet. In other environments, it can be combined with extant probing algorithms to further improve performance. © , the latency of £ ¤ is © , and the latency of ¤ is © . Then we have the following linear equations: