IPv6 was designed to cater for many deployment scenarios, starting with extension of the packet technology and therefore supporting IPv4 with transition models to keep IPv4 working even for ever and then to cater for new uses and new models that require a combination of features that were not tightly designed or scalable in IPv4 like end to end connectivity, end to end services, ad hoc services, end to end QoS, end to end security, auto-config on the fly, large-scale multicast, anycast services and IP mobility; to the extreme scenario where IP becomes a commodity service enabling lowest cost deployment of large scale cloud computing, Internet of Things, SmartGRIDs, Green IT, sensor networks, RFID, IP in the car, to any imaginable scenario where networking adds value to commodity. The network management and the maintenance of the infrastructure become a central and critical element and here again IPv6 extends reachability, remote access and remote diagnostics at node level. The autonomicity takes a new dimension with use of IPv6 extensions, there's a potential to open new frontiers in this research area. This is called progress in networking making IP the dominant open Internet protocol.

The performance of mobile ad hoc network is influenced by a number of factors, such as protocol selections, parameter settings, wireless channel conditions, network size, transmission ranges, and traffic loads. In this work, we present an Autonomic Network Performance Management (ANPM) framework for ad hoc networks. The proposed framework follows a control-theoretic approach and is designed to monitor, model, optimize, and configure the controllable node and system-level factors to satisfy global performance goals. The functional components of the ANPM framework are presented along with a case study of establishing a call center in a disaster recovery operation using MANET. Tuning the performance of the overall network performance is shown through: (1) changing whole network protocols, or (2) changing within the network protocols.

Future Long Term Evolution (LTE) Radio Access Network (RAN) will benefit from a significant degree of self-organization. Autonomic Load Balancing (ALB) is considered as an important self-organizing function in LTE RAN. In this work, distributed ALB is achieved by Autonomic Network Management (ANM). Load conditions of eNBs (E-UTRAN NodeB) can be detected. Sizes and shapes of cellular coverage can be adjusted automatically according to load conditions, so as to balance load. Simulation results demonstrated that by the use of ALB, the system capacity can be improved significantly.

Search efficiency and accuracy of resource are important considerations in P2P networks. Most researches focus on improving search efficiency through efficient search mechanisms. However, they ignore the impact of P2P network topology on search efficiency, especially in heterogeneous network environments. In this paper, we propose a novel semantic topology construction method CATS （Capacity-Aware Topology based on Semantic） based on Vector Space Model (VSM). This topology construction method considers node capacity based on semantic in heterogeneous environment. Through building a rational and effective semantic topology, it significantly improves the search efficiency. Meanwhile, we adopt a novel and efficient search mechanism BF-SKIP (Biased walk, Flooding and Search with K-Iteration Preference) to promote searching within network. Compared with the current homogeneous semantic-based topology construction method GES, CATS outperform GES in terms of higher recall and lower search cost.

This paper introduces a hierarchical trick play mechanism for video streaming. Trick level concept is introduced to support multiple trick rates, reduce delay when switching from high to low rate and minimize network bandwidth usage. Dedicated trick play streams on server side which are created hierarchically are used to improve user experience and reduces client processing requirement. It is efficient and easy to implementation.

Autonomic networks are an emerging technology which is promising to reduce the complexity of human-driven network management processes and enable a variety of so-called self-* features such as self-configuration, self-optimization, etc, inside the network devices and the network as a whole. Autonomic behaviors are widely understood as a control loop implemented by an autonomic entity that automates management processes and controls diverse aspects of a set of resources. Though automation is necessary and achievable, autonomic decision-making-elements of the network can not fully perform decisions on every task of the network without requiring some degree of "a human-in-the-loop" in some of the decisions. From the operator's perspective, controllability of the control loop and decision notification from the autonomic network is a vital issue that needs to be addressed. In this paper we present our considerations on how an Autonomic Fault-Management control loop ("detect an incident" - "find the root cause behind it" - "remove the root cause") can be controlled by the network operation personnel.

Online configuration of large-scale systems such as networks require parameter optimization to be done within a limited amount of time. This time limit is even more pressing when configuration is needed as a recovery response to a failure in the system. To quickly configure such systems in an online manner, we propose a Probabilistic Trans-Algorithmic Search (PTAS) framework which leverages multiple optimization search algorithms in an iterative manner. Essentially, PTAS applies a search algorithm to find out how to best distribute available experiment budget among multiple optimization search algorithms. Specifically, PTAS allocates experiment budget to each available search algorithm and observes each algorithm's performance on the system-at-hand. PTAS then probabilistically reallocates the experiment budget for the next round proportional to an algorithm's performance. This "roulette wheel" approach probabilistically favors the more successful algorithm in the next round. Following each round, the PTAS framework transfers the best found result(s) among the individual algorithms, making our our framework "trans-algorithmic". PTAS thus aims to systematize how to "search for the best search". We show the performance of PTAS on well-known benchmark objective functions including scenarios where the objective function changes in the middle of the optimization process. To illustrate applicability of our framework to automated network management, we apply PTAS on the problem of optimizing link weights of an intra-domain routing protocol on a topology obtained from Rocketfuel dataset.

Carrier networks need to provide their customers with high availability of communications services. Unfortunately, failures are managed by recovery mechanisms getting involved only after the failure occurrence to limit the impact on traffic flows. However, there are often forewarning signs that a network device will stop working properly. We propose to take into account this risk exposure in order to improve the performance of the existing restoration mechanisms, in particular for IP networks. Based on an embedded and real-time risk-level assessment, we can perform a proactive fault-management and isolate the failing routers out of the routed topology, and thus totally avoid service unavailability. Our novel approach enables routers to preventively steer traffic away from risky paths by temporally tuning OSPF link cost.

In general, network traffic is classified and marked at the edge- or ingress-routers of a transit network to permit different levels of QoS when using the DiffServ architecture. Thus, the core-nodes of a network are able to handle the traffic accordingly. Problems in the process of identification of traffic flows and their associated applications, for example because of encryption, avoid the classification of network traffic. This can be improved by relocating the functions of classifying and marking towards the network traffic generating hosts. The great benefit of the relocation is that applications can be identified in any case. In addition, a pre-shaping of network traffic becomes possible. Unfortunately, the set of rules for classification and marking on hosts is always static in its behavior so that it would not be possible to map dynamics, for example, of business processes to the set of rules or its interpreter. Therefore, it has to be possible that the marking of packets depends on their temporal execution context and, thereby, on the business processes. In this paper, we present an efficient solution that considers business processes in host-based QoS management systems. The applied QoS management system considers the dynamic requirements of business processes and works application-independent. The performance of the system, introduced by its efficient design, is analyzed and its benefits are discussed.

The success of future generation networks will be driven by the provision of ubiquitous, personalized services that can offer an enhanced user experience. In order to satisfy these business requirements, for the provision of more and better services to users, and at the same time deal with the complexity of the infrastructure of network operators (NOs), advanced management functionality needs to be introduced in wireless systems, which will enable optimum, end-to-end operation. Numerous schemes for co-operative, adaptive, reconfigurable and cognitive systems have been developed as a means for addressing many of the requirements of future networks. However, the vast majority of the corresponding implemented platforms and test beds are proprietary, custom solutions that focus on some aspects of cognitive systems, e.g. on dynamic spectrum management capabilities. A framework, that can facilitate the integration of existing solutions and of new, emerging cognitive schemes, so as to reach the full potential of future networks, is missing. The contribution of this paper is in this direction. This paper presents a platform for the integration and management of cognitive systems (IMaCS) that aims at smoothing the progress of introducing various cognitive schemes in future networks. More specifically, building on past achievements the platform presented in this paper not only comprises various cognitive management schemes, thus enabling efficient end-to-end operation, but also targets the abstraction of the complexity of the underlying infrastructure. Furthermore, a service-oriented design approach has been followed with the goal of realizing an open, scalable platform that will allow the addition, removal or enhancement of components (cognitive management schemes, devices, network elements) in a "plug and play" manner. The features of the main IMaCS components are outlined. The information flow between these components is described in detail. An overview of the corresponding IMaCS platform implementation is provided. The paper also presents indicative results on the performance of the IMaCS platform, in terms of bytes and time required for the exchange of messages between the various components.

The increasing size, complexity and the dynamic character of future networks make traditional monitoring systems inadequate to be continuously updated and sense the endless changes in the topology and communication conditions. The need for reduction in the network management complexity and the administrator's operational burden imposes the design and implementation of self-functionalities and the adoption of self-management schemes. We propose a novel approach for improving network management based on distributed monitoring techniques. Autonomic mechanisms are designed to control traffic monitoring within a network in order to increase the accuracy of the acquired data without increasing the imposed overhead in the network. Nodes are able to correlate monitoring data and to adapt the monitoring functions according to existing networking conditions and predefined administration policies.

The addressing and forwarding architecture based on the destination of packets in current Internet typically does not check the authenticity of source address of packets; therefore, it causes a considerable challenge to prevent the attackers from launching attacks by forging source addresses and to trace the real sources which sent the malicious traffic. In this paper, we present a new protocol/architecture designed to enhance network security by separately verifying the authenticity of source address in the ingress of access network and the credibility of packet path on the border of every domain. The access validation bases on the label generated by host; the packet authentication is implemented by an indicator of accumulated information of domains which the packets pass through. This mechanism intrinsically provides the IP traceback. This paper also discusses the issues of development and deployment in practice, and two prototype drafts are given.

Reciprocity-based cooperation on packet relaying in mobile ad hoc networks discourages nodes from behaving selfishly. Before passing on a packet to the next hop a node verifies whether the sender of the packet is trustworthy (cooperative) or not. The quality of the evaluation of trustworthiness relies on types of data used for description of the behaviour of the sender. In the literature a distinction was made between first-hand observations and observations obtained from third parties. This paper demonstrates that further distinction of such data into personal and general can help to create better protection against clique building among nodes. The former takes into account status of packets originated by a node itself, while in the latter the status of packets originated by other nodes is considered. Computational experiments demonstrate that in the presence of a large number of selfish and colluding nodes prioritizing the personal data improves the performance of cooperative nodes and creates a better defence against colluding free-riders.

Grid computing aims at offering standardized access to heterogeneous and distributed resources for scientific communities. However, in order to support emerging next generation Grid applications with specific Quality of Service requirements, the interconnecting networks have also been considered as first-class allocable Grid resources and have been also taken into account for the co-scheduling process. In the last few years, a number of network resource provisioning system were developed, however, without providing specific analysis on the scalability of potential architectural design alternatives. Our approach is to formulate a fundamental analytical model to evaluate the expected service provisioning time using different architectures as a function of the involved transport domains. To validate our results, we have used measurements obtained from the European IST-FP6 Phosphorus project testbed. The main contribution is to provide an instrument to obtain reference values to support architectural design decisions even in an early stage of the development phase.

The proliferation of large-scale P2P VoD overlays has created the demands to characterize and to understand the emergent properties of P2P VoD applications. In this paper, we investigate the P2P VoD overlay network by active measurements with the help of the developed software, VoDCrawler. Different from other existing work, our study focuses on the topology of the overlay networks. From the study, we have revealed the major features of the P2P VoD overlay networks and have compared them with that in P2P file sharing and live streaming systems. The discoveries we have obtained could help to further understand the operation of the P2P VoD systems and favor the commercial developments of the P2P VoD applications.

Nowadays energy management is a key feature in many different fields, specially in mobile ad hoc networks where devices heavily rely on the battery life, thus, the network survival is absolutely related to the energy consumption of nodes. In this work, we present a broadcasting algorithm AEDB that not only tries to reduce the network but also the device resources. AEDB is an extension of EDB which is a distance based broadcasting and also energy aware. The new proposed scheme, AEDB, regulates the transmission power of the device in order to decrease the energy consumption with no detriment of the performance of the algorithm. This is done by managing the transmission power of the device in terms of the number of one hop neighbors for disseminating a message. Results show that by adaptively managing the transmission power, the energy the device uses is highly reduced obtaining at the same time, good performance in terms of both the coverage achieved and the broadcast time.

To predict the up/down state of a node in P2P network, its history trace is usually the information source, from which the user behavior is commonly investigated. However, if the information of a single node is not sufficient enough, some usage patterns cannot be captured precisely, which may lead to a lot of predicting errors. In this paper, we designed an approach to cluster the nodes based on their similar behaviors, and the all information in a same cluster is seen as another information source for any node in it. Therefore, a novel predictor trained by a combination of different information sources is proposed. In our approach, N-gram model is used as the basic predictor. Evaluation shows that if a node's trace is enhanced by the cluster information, the performance of the predictor trained based on the single source can be significantly improved, and this sources combined predictor obviously outperforms the state-of-the-art system, whose relative error reductions in different observation time are about 0.1%~36.3, 0.6%~20.1% and 6.1%~44.5% for Overnet, Skype and Microsoft datasets respectively.

The talk will offer some perspectives on how to create a viable Evolution Path towards Self-Managing Future Internet via the use of a standardizable, evolvable and commonly-shared architectural Reference Model for Autonomic Network Engineering and Self-Management for Future Network design. We present a scenario on how the Self-Managing Future Internet can emerge through a viable Evolution Path that considers incrementally evolving today's network models, architectures, protocols such as IPv6 (in particular) and paradigms. The scenario goes on to define the incremental changes and concepts necessitated and guided by a unified, holistic, commonly-shared architectural Reference Model for Autonomic Network Engineering and Self-Management that needs to be established first as a starting point to creating the Evolution Path towards the Self-Managing Future Internet. Significant efforts are now being invested towards establishing a Generic Autonomic Networking Architecture (GANA) as Reference Model for Autonomic Network Engineering and Self-Management for Future Networks. As a Reference Model, it is meant to establish common language, understanding and reasoning about the design principles for autonomic management and control of managed entities (both, functional entities and physical resources) by Decision-Making-Elements. It must also reflect and describe hierarchies and horizontal peer relations and interaction-flow in decision-making and management. Also to be reflected in the Reference Model, are the Decision-Making-Elements: (1) as inter-working drivers of hierarchical control-loops and as containers of cognitive and learning algorithms, and (2) their levels of operation within the architecture of a node/device up to the level of the network architecture as a whole. This calls for standardization, since no commonly shared holistic Reference Model existed before. The evolution of today's network models, architectures, networking paradigms and protocols such as IPv6 (towards IPv6++) must be guided and necessitated by this type of the sought architectural Reference Model. The Scenario is a "what-if" type of Scenario that presents solid and realistic steps that define an evolutionary roadmap to achieving a very advanced feature-rich Self-Managing Future Internet by 2015 (or possibly earlier), which can continue to evolve beyond that time frame. On the other hand, now is the time to digest and bring the good and validated concepts currently scattered in conference/workshop/journal publications, into the development of the architectural Reference Model. Such validated concepts as they make their way into the Reference Model, will also be applied in the incremental evolution of today's network models, architectures, protocols such as IPv6, as the Reference Model is instantiated for autonomic management and control of today's technologies and protocols in diverse network environments and contexts. Concrete implementation architectures derive from an instantiation of the architectural Reference Model for a particular type of network environment, device roles and context.

In recent years, the concept of autonomicity has been proposed to solve the complexity issues of the network and reduce the manual intervention. The self-* properties of the autonomic network such as self-configuration, self-healing, self-optimizing and self-management enable the network to adapting the heterogeneous and complex network environments. On the other hand, with the rapidly growing demand for Quality of Service (QoS), it is hard to satisfy the QoS requirement with the traditional manual methods. In this paper, a novel autonomic architecture for QoS management in wired network is proposed. With a clean-slate A4D (Autonomic 4D) approach adopted and autonomic mechanisms introduced into the traditional DiffServ framework, the proposed architecture is capable of adjusting network behavior adaptively to meet the users' QoS requirements.

one of the significant problems that high-tech companies are facing is the management and monitoring of networks in order to provide better and more reliable services for their customers. This paper introduces a new approach for the prediction of network failure and performance degradation using Joint Clustering and Association Analysis approach (JCAA). JCAA differs from existing prediction techniques in terms of exploiting the clustering and association analysis techniques in order to improve the quality of prediction. The role of clustering is to classify the input data into groups of k-means clusters, while the association analysis technique discovers the causal relationships between the groups. The experimental results demonstrate that the proposed system is truly effective in enhancing the quality of prediction.

The use of cellular networks for both voice and data communications is growing at a rapid pace throughout the world. Such growth was initially fueled by the need for mobile voice communication. However, with the advent of 2G networks and now with 3G networks, data services are quickly outpacing that of voice. Normally such growth is welcomed since it results in increased revenue for the wireless operator. However, unlike wireline networks for which capacity can easily be increased, the capacity of wireless networks is limited. The load that can be sustained by a wireless network depends on various factors but the major ones are cell size, bandwidth and spectral efficiency. Each of these have limits and so as the demand for data services increases, the pricing and allocation of the limited available resources become increasingly more important. In this paper we address the impact that limited resources have on pricing and how, in turn, radio resource management algorithms are affected. The emphasis in this work is on the overall framework aspects and not on the designs of the individual radio resource management algorithms.

This paper presents the key designing principles and attributes of a novel autonomic architecture for enabling efficient mobility and QoS-aware resource management over an integrated IPv6-enabled heterogeneous wireless environment (CDMA cellular/WLAN). The proposed architecture is founded on Generic Autonomic Network Architecture (GANA), a generic framework for realizing autonomic networks introduced in EFIPSANS EC FP7 project. Emphasis is placed on describing the autonomic components of the architecture, illustrating their interactions and justifying their designing properties towards creating a concrete autonomic paradigm that aims at fulfilling 3GPP/LTE self-optimization vision.

In this paper, we offer some perspectives on how to create a viable Evolution Path towards Self-Managing Future Internet via a standardizable and commonly-shared architectural Reference Model for Autonomic Network Engineering and Self-Management. We present a Scenario on how the Self-Managing Future Internet can be developed via a viable Evolution Path that starts with today's network models, architectures, protocols such as IPv6 (in particular) and paradigms. The scenario then goes on to define the incremental changes and concepts necessitated and guided by a unified, holistic, architectural Reference Model for Autonomic Network Engineering and Self-Management that needs to be developed first, as a starting point to creating the Evolution Path towards the Self-Managing Future Internet. This evolution of today's network models, architectures, networking paradigms and protocols such as IPv6 (towards IPv6++) must be guided and necessitated by the architectural Reference Model. The Scenario is a "what-if" type of Scenario that presents solid and realistic steps that define an evolutionary roadmap to achieving a very advanced feature-rich Self-Managing Future Internet by 2015, which can continue to evolve beyond that time frame. The ongoing activities of the EC funded FP7-EFIPSANS Project (http://www.efipsans.org/ ) are geared towards this goal.

Network elements for constructing an IP network have been increasing in number with the expansion of the telecommunications infrastructure. Network subscriber services are enabled or disabled by setting command lines into network elements. Moreover, the diversification of network services and the increase in the number of subscribers are two factors that increase a network management system's ability to input command lines to network devices. Therefore, to manage a network more efficiently, the importance of setting command lines into network devices more efficiently is growing. We are developing a high-speed task processing method considering parallel-session status for systems remotely controlling several routers. We show the results of simulations and propose another two merging and optimizing methods for command scenarios of setting network service orders into the same network element and the same VLAN interface of a network element. These three methods are used to shorten session occupancy time per device to achieve the maximum effect with limited session resources.

``Give a man a fish, feed him for a day. Teach a man to fish, feed him for a lifetime'' -- Lau Tzu Grid computing projects such as TeraGrid, Grid'5000, and OpenScience Grid enable researchers to access vast amounts of compute resources, but in doing so, they force the researcher to adapt their workloads to the environments these systems provide. This does not leave many alternatives for researchers as creating these types of systems requires coordination and expertise in networking, operating systems, security, and grid middleware. This results in many research groups creating small, in-house compute clusters where scheduling is often ad-hoc (and resource utilization is low), and aggregation of resources across multiple groups is hindered by the complexities in constructing federated systems. This paper describes the ``Grid Appliance'', the first system of its kind that enables researchers to efficiently deploy their own compute clusters, and to seamlessly extend their systems across network domains to create small to large scale computing grids. The paper details the design of the Grid Appliance and reports on experiences and lessons learned over four years of development and deployment of wide-area Grid appliance pools.

Dynamic allocation of multiple cloud resources adapting to applications needs over time can be achieved by taking advantage of wide-area VM live migration technologies. However, migration of VMs across different subnets, potentially in multiple clouds, requires networking support to keep the network state of moving VMs unchanged. Two problems make traditional solutions to machine mobility inefficient in this scenario: (1) administrative overheads due to coordination requirements between moving machines and the network infrastructure; and (2) degraded network performance of machines moved away from their "home" networks. New solutions are needed to efficiently support the migration of virtual machines over multiple cloud providers. The user-level virtual network architecture presented in this paper implements mechanisms to allow VM migration over clouds without requiring support from the physical network infrastructure, and automatically reconfiguring virtual networks to maximize the network performance of migrated virtual machines.

This paper describes the experience of developing a system for creation of distributed linked configurations of heterogeneous resources (slices) in GENI. Our work leverages a number of unique architectural solutions (distributed architecture, declarative resource specifications, unique approach to slice instantiation) which is applicable to a wider set of problems related to autonomic co-scheduling and provisioning of heterogeneous networked resources. We discuss the architecture, the resource description mechanisms and some of the algorithms used to enable our system. We conclude with an analysis of a real experiment at allocating resources from multiple providers across a very wide geographic area (spanning Massachusetts, Illinois and North Carolina) to create a single private Layer 2 network connecting virtual machines on the campus of Duke University to a sensor testbed at University of Massachusetts, Amherst.

In this paper we face the challenging issue of defining and implementing an effective law for load balancing in Content Delivery Networks. We base our proposal on a formal study of a CDN system, carried out through the exploitation of a fluid flow model characterization of the network of servers. Starting from such characterization, we derive and prove a lemma about the network queues equilibrium. This result is then leveraged in order to devise a novel distributed algorithm for load balancing. The overall approach is validated by means of simulations showing the effectiveness of the proposed algorithm in terms of both fair load distribution and limited service time.

This paper discusses a novel data center architecture based on in-packet Bloom filters to perform valiant load balanced packet forwarding and two support services do distribute the directory and topology state of OpenFlow controllers. By deploying an army of Rack Managers acting as OpenFlow controllers, the proposed architecture promises scalability, performance and fault-tolerance. We conjecture that packet forwarding itself may become a cloud internal service implemented by leveraging cloud application best practices such as distributed key-value storage systems. Moreover, we contribute to demystifying the argument that the centralized controller model of OpenFlow networks is prone to a single point of failure and show that direct network controllers can be physically distributed, yielding thereby a sweet networking approach between fully distributed and centralized.

We are investigating new methods for Configuration Management of devices inter-connected over challenged networks such as wireless Mobile Ad-Hoc Networks (MANETs) and Disruption Tolerant Networks (DTNs). Configuration Management in these network environments require efficiency in communications, bundling of base operations into composite sets, i.e., transactions, resiliency against configuration errors and local autonomous control of configuration. This paper describes our early efforts in architecture, designing and prototyping the Configuration Network Management Protocol (CNMP) as a proposed Configuration Management protocol for these challenging network environments.

This paper introduces a platform for realistic and computationally efficient online vehicular networks simulation. It permits decentralized traffic management applications simulation as nodes mobility is modifiable at runtime thanks to the integration of two state-of-the-art network and a traffic simulators. The platform additionally embeds a  micro-simulation tool that generates mobility patterns based on traffic counting data and ensures performance through a geographical decomposition of the network. Evidence of its performance is given on a Luxembougian traffic management scenario, using real road network and traffic data.

In this paper we discuss the viability of utility-based abstractions for network management. These can be used for estimation and optimization of the network performance. We consider utilities that are determined by the application type, but also take into account changing user and provider context. These metrics incorporate technical performance characteristics, economic and personal considerations of network stakeholders. We discuss on approximations of utility functions and related optimizers. Utility approximations are particularly attractive option to avoid the considerable signaling and computational overhead that otherwise can be imposed by overly detailed specification of optimization problems. As examples we consider simplified versions of the class-based advanced queue management, and the workload/cost balancing between data centers. Our experiments show that in some cases up to 30\% performance gains can be achieved compared to the standard approaches.

This paper proposes a novel admission control (AC) algorithm for guaranteed quality of service (QoS) to all users. The proposed solution provides better utilization of system capacity using adaptive modulation (AM). A scheduler based on a per user priority function is also given in this paper. The AC is given by allocating the exact number of slots for each user that will meet its QoS. For every new user the number of slots required to meet its service requirements is estimated based on its channel quality information (CQI), packet arrival rate and buffer length. Using the average QoS achieved, the satisfaction index (SI) and priority is calculated for every user, which is used as key input for the scheduler. Further, the resource allocation in time and frequency for Orthogonal Frequency Division Multiplexing (OFDMA) systems is also discussed based on user satisfaction and number of slots required for each user. Finally, the proposed design is validated with OFDMA systems, but can be extended to any wireless system.

Flow-aware networking (FAN) is an alternative QoS mechanism proposed in the context of an unsatisfactory solution to QoS provision for real-time traffic. It categorizes the Internet traffic into stream flows and elastic flows and realizes service differentiation between them. Despite some studies in wired networks, the performance of FAN has rarely been analyzed in wireless networks, which are gaining great popularity in recent years. In this paper, we took the first step to design a framework for the performance analysis of FAN in wired-cum-wireless networks using stochastic network calculus. Different from most studies which focused on the deterministic service guarantees provided by FAN, this paper shed light on the stochastic analysis of its achievable level of performance.

A heterogeneous wireless network supporting multihoming gives multi-mode terminals the flexibility to be simultaneously connected to more than one radio access technologies (RATs). Existing joint call admission control (JCAC) algorithms designed for heterogeneous wireless networks block or drop an incoming call when none of the available individual RATs in the heterogeneous network has enough bandwidth to support the incoming call. Consequently, high bandwidth-demanding calls can easily be blocked or dropped in the network, especially during the peak hours. In order to reduce this problem of call blocking/dropping, this paper proposes a JCAC algorithm that selects multiple RATs for an incoming call when none of the available individual RATs has enough bbu to accommodate the incoming call. Selection of multiple RATs for an incoming call entails that the packet stream of the incoming call will be split among the selected RATs. The aim of the proposed JCAC algorithm is to admit an incoming call (that cannot be admitted into any of the available single RATs because of high load in the RATs) into two or more RATs. The residual bandwidths in the selected RATs are combined to support the incoming call, and the packet stream of the call is split among the selected RATs, thereby reducing call blocking/dropping probability. At the receiver, the split packet streams are then combined. An analytical model is developed for the proposed JCAC algorithm, and its performance is evaluated in terms of call blocking/dropping probability. Simulation results show that the JCAC algorithm reduces call blocking/dropping probability in heterogeneous wireless networks supporting multihoming.

The proliferation and integration of communication networks in social life has increased the need for trusted systems of advanced and intelligent capabilities. Future networks are calling for new ways to efficient management, operation and service provisioning. Autonomicity becomes an enabler for self-manageability of future networks and therefore autonomic networking provides the necessary new paradigm for these networks to become manageable and scalable. Autonomic entities base their decision within a network on experience gathered and information exchanged. Trust management mechanisms can provide the necessary security framework in such an environment towards robust coherent autonomic networking. In the paper we present trust models and sketch a trust management architecture, applicable to complex future networking environments. We handle the special requirements set by autonomicity and try to strengthen the autonomic characteristics of the nodes as well as the robustness of service provisioning.

Large enterprises usually require Virtual Private Network (VPN) services provisioned by the network operator. Also, there is an emerging need for supporting multicast communications, i.e. one host communicate with other hosts located in multiple remote sites. While MPLS-based IP VPNs are proven to be scalable, current approaches for extending it with multicast features involve potential state explosion, some bandwidth inefficiencies in the operator network or complex management tasks to find a good balance between forwarding state and bandwidth usage. These properties are direct consequences of the current MPLS and network-layer multicast forwarding approaches, where state should be maintained in the forwarding plane for each tree in each intermediate node. In this paper, we build on a stateless Bloom-filter-based forwarding plane installed in the service provider's network. By moving the state into the packet headers from the nodes, new trade-offs appear due to the probabilistic nature of Bloom filters. We highlight autonomic scenarios, such as self-configuration of addresses, resource management in the network, simple autonomic provisioning of dynamic multicast trees and self-optimization of forwarding performance.

In view of the adaptive Quality of Service (QoS) maintenance issue for multi-tire virtualized IT Service, basing IT resource regulation only on QoS reference value leads to suboptimal decision. In this paper, we propose an online QoS maintenance mechanism for multi-tier IT service system in order to obtain the optimal global business utility. We build a novel business utility model which concerns both the business revenue and loss related to service performance and availability for multi-tier virtualized IT service. Then we establish the resource allocation mechanism with the assistance of our proposed novel IT-business metrics mapping. This approach can adaptively maintain the QoS in a reasonable range by dynamically regulating the virtualized IT resource. The experimental simulation shows that our approach is superior to the conventional approach since the overall business utility of all service levels stays on a relative high level.

Service-oriented communication is a new trend in the industry, which opens a new paradigm of Web service based VoIP communication and can be easily integrated in end-to-end SOA solutions. In this paper, we present the real-time communication components based multimedia conferencing Web service over converged networks, and mainly focus on the design and implementation of JAIN-SLEE service build block (SBB) based real-time communication components, the corresponding SIP signal control flows, session management and state machine, and also describe the implementation for SBB based valued added service to multimedia conferencing communication Web service. And also a prototype has been implemented and the performance has been measured. Finally, give the conclusions and future work.

Nowadays we are witnessing a strong interest from the network operators towards provisioning all-IP services. The user demands and the effort of the operators to answer them push the networks to a complexity where traditional network management methods do not scale. Operators try to introduce new functionalities and mechanisms that allow the network to be more independent from human intervention. Autonomic networking is an answer to these complexity issues related to network management. Within the EFIPSANS project, funded by EU, we have proposed the Generic Autonomic Network Architecture (GANA), which offers a modular solution to such issues. Present paper presents the GANA, a step on the path towards the realization of autonomic networks. We also present the integrated EFIPSANS testbed and the implemented mechanisms, designed to demonstrate a substantial selection of essential autonomic behaviours.

Protocols such as 6LoWPAN will soon directly connect the Internet to the rapidly growing number of deployed Low Power and Lossy Networks (L2Ns). The outcome will enable many new applications, including new types of Cyber-Physical Systems, the "Internet-of-Things", the Smart Grid, etc. Due to resource constraints the performance of a L2N routing protocol is quite sensitive to changes in protocol parameters such as route maintenance and packet retransmission timers. However, the lack of network management tools makes it quite difficult to efficiently set these parameters or even to efficiently compare different protocols across multiple metrics of reliability, energy consumption, throughput and delay. Our paper addresses this issue. We discuss the most important parameters that must be managed within an Internet-friendly L2N. We propose a novel management heuristic called ROLL-POWER, designed to simplify the monitoring of L2N performance, independent of the routing protocol type. Through extensive simulation we evaluate our claims about the importance of correct parameter setting and the efficiency of the ROLL-POWER metric. Our results show that this simple metric can be used by a network manager both to tune protocol parameters for L2Ns and as a simple and direct method of comparing different protocols.