Serverless computing is a recently introduced deployment model to provide cloud services. The autoscaling of function instances allows adapting allocated resources to workload, so as to reduce latency and improve resource usage efficiency. However, autoscaling mechanisms could be affected by undesired ‘cold starts’ events, causing latency peaks due to spawning of new instances, which can be critical in edge deployments where applications are typically sensitive to latency. In order to regulate autoscaling of functions and mitigate the latency for accessing services, which may hinder the adoption of the serverless model in edge computing, we resort to the usage of reinforcement learning. Our experimental system is based on OpenFaaS, the most popular open-source Kubernetes-based serverless platform. In this system, we introduce a Q-Learning (QL) agent to dynamically configure the Kubernetes Horizontal Pod Autoscaler (HPA). This is accomplished via a QL model state space and a reward function definition that enforce service level agreement (SLA) compliance, in terms of latency, without allocating excessive resources. The agent is trained and tested using real serverless function invocation patterns, made available by Microsoft Azure. The experimental results show the benefits provided by the proposed solution over state-of-the-art in terms of compliance to the SLA, while limiting resource consumption and service request losses.

Modern cloud and edge computing infrastructures operate under rigorous Service Level Agreements (SLAs), which define Quality of Service (QoS) standards and key performance indicators (KPIs) such as latency, resource efficiency, and reliability. Meeting these requirements is increasingly challenging due to highly variable, bursty, and latency-sensitive workloads demanding intelligent, efficient, and adaptive resource management. To address these evolving needs, we introduce a novel autoscaling framework that combines the Proximal Policy Optimization (PPO) algorithm with a double-stacked Long Short-Term Memory (LSTM) network and an attention mechanism, enabling coordinated and proactive autoscaling in Kubernetes-based environments. The proposed approach utilizes a multidimensional discrete action space to simultaneously tune the Horizontal Pod Autoscaler (HPA) CPU targets, throughput multipliers, learning rate schedules, and policy enhancements, thereby effectively adapting to both short-term and long-term fluctuations in network traffic. Additionally, a forecasting module integrates LSTM-driven workload predictions with recent observation trends, allowing the autoscaler to predict dynamic changes and accommodate resources accordingly. Comprehensive simulations using real Azure Functions invocation traces from production environments justify that our system consistently maintains SLA, QoS, and KPI objectives, while delivering operational stability and enhanced resource efficiency for cloud-edge networks. Under strict SLA constraints, our proposed methodology achieves an average latency reduction of $\mathbf{8 7 . 5 \%}$ and a median latency reduction of $\mathbf{7 1 . 5 \%}$ compared to the baseline.

Serverless computing is an emerging service model that is gaining consensus in both edge and core clouds. In fact, it allows developers focusing on application codes, breaking down services in multiple functions, without the burden of platform management. Services are run in edge clouds, closer to end-users’ devices, when service latency is an important constraint. However, latency control cannot be achieved with ample resource allocation, as in edge environments resources are limited and thus operational efficiency is essential. For these reasons, autoscaling of serverless functions has a central role, as it can lead to both reduced latency and efficient resource usage. In this work, we focus on the usage of reinforcement learning to drive resource-based autoscaling on OpenFaaS, the most-adopted open-source serverless platform, running on Kubernetes clusters. We use the Proximal Policy Optimization algorithm to dynamically configure the value of the Kubernetes Horizontal Pod Autoscaler, trained on Azure traffic traces. We discuss the algorithm configuration, focusing on state definition and relevant performance evaluation, including robustness with respect to traffic burstiness, as well as performance comparison with baseline approach and heuristics.

The service function (SF) area has gained increasing attention in the last years due its ability to combine the advantages of cloud computing with network softwarization. By decoupling SFs from the physical equipment where they are executed, it is possible to make network services scalable and flexible. These advantages become even more evident in the forthcoming 6G networks, where the overall environment is expected to become more dynamic and cloud-based, with SFs deployed as cloud-native functions. However, in order to efficiently manage and compose services using these SFs, it is necessary to monitor the available resources of the nodes where they can be deployed, in addition to exchange information relevant to the operational status of active SFs. To this aim, we propose a lightweight monitoring architecture by using agents in charge of monitoring the status of SFs running in co-located clusters. These monitoring agents exchange their information by means of a gossip protocol, which allows increasing the reliability of the process. In this way, it is possible to keep service decisions as local as possible, limiting the interactions with centralized decision and orchestration platforms, and thus increasing network scalability and responsiveness. Performance evaluation shows the effectiveness of the proposed solution, and demonstrates that the network overhead of the distributed monitoring process is definitely affordable.

Fifth generation (5G) and beyond systems require flexible and efficient monitoring platforms to guarantee optimal key performance indicators (KPIs) in various scenarios. Their applicability in Edge computing environments requires lightweight monitoring solutions. This work evaluates different candidate technologies to implement a monitoring platform for 5G and beyond systems in these environments. For monitoring data plane technologies, we evaluate different virtualization technologies, including bare metal servers, virtual machines, and orchestrated containers. We show that containers not only offer superior flexibility and deployment agility, but also allow obtaining better throughput and latency. In addition, we explore the suitability of the Function-as-a-Service (FaaS) serverless paradigm for deploying the functions used to manage the monitoring platform. This is motivated by the event oriented nature of those functions, designed to set up the monitoring infrastructure for newly created services. When the FaaS warm start mode is used, the platform gives users the perception of resources that are always available. When a cold start mode is used, containers running the application’s modules are automatically destroyed when the application is not in use. Our analysis compares both of them with the standard deployment of microservices. The experimental results show that the cold start mode produces a significant latency increase, along with potential instabilities. For this reason, its usage is not recommended despite the potential savings of computing resources. Conversely, when the warm start mode is used for executing configuration tasks of monitoring infrastructure, it can provide similar execution times to a microservice-based deployment. In addition, the FaaS approach significantly simplifies the code logic in comparison with microservices, reducing lines of code to less than 38%, thus reducing development time. Thus, FaaS in warm start mode represents the best candidate technology to implements such management functions.

5G technology offers innovative network services to vertical players, able to boost their service portfolio in critical areas that could not be fully addressed with 4G systems. However, 5G networks are complex systems, massively relying on the virtualization of network and computing resources, which requires specific skills to fully exploit their potential. This may represent a hard barrier for vertical stakeholders to effectively test and use innovative services on 5G networks. Our solution addresses this issue, by simplifying the way vertical players approach 5G services though simple yet flexible service templates that allow not only to specify and deploy a service, but also to test it in a variety of realistic conditions. Vertical players interact with these templates named Experiment Blueprints, which are manipulated by our tool and then translated into network service descriptors to be deployed by a service orchestrator, keeping unaware the vertical experimenter of the underlying operations. We use a vertical service in the area of Industry 4.0 as a running example to ease the understanding of the proposed approach.

Due to the growing number of devices accessing the Internet through wireless networks, the radio spectrum has become a highly contended resource. The availability of low cost radio spectrum monitoring sensors enables a geographically distributed, real-time observation of the spectrum to spot inefficiencies and to develop new strategies for its utilization. The potentially large number of sensors to be deployed and the intrinsic nature of data make this task a Big Data problem. In this work we design, implement, and validate a hardware and software architecture for wideband radio spectrum monitoring inspired to the Lambda architecture. This system offers Spectrum Sensing as a Service to let end users easily access and process radio spectrum data. To minimize the latency of services offered by the platform, we fine tune the data processing chain. From the analysis of sensor data characteristics, we design the data models for MongoDB and Cassandra, two popular NoSQL databases. A MapReduce job for spectrum visualization has been developed to show the potential of our approach and to identify the challenges in processing spectrum sensor data. We experimentally evaluate and compare the performance of the two databases in terms of application processing time for different types of queries applied on data streams with heterogeneous generation rate. Our experiments show that Cassandra outperforms MongoDB in most cases, with some exceptions depending on data stream rate.

This paper shows a codebook-based proposal for identifying simultaneous faults in data networks. It includes three main contributions. The first consists of the probabilitstic reduced search space heuristic (PRSSH), which aims to significantly reduce the cardinality of the candidate set of solutions through a so-called compatibility filtering. It is applied to the codebook optimized by the second contribution, which is a solution of the codebook optimization problem. It is built on the weighted set covering optimization problem and consists of a heuristic, named minimum Hamming distance increment maximization heuristic (MHDIM-HEU). Performance has been evaluated both by applying the proposed techniques to some sample networks, simulated by using the end-to-end connectivity service model, and by using an experimental codebook generated by using data collected from a real, nation-wide, next generation network. This codebook has been used to simulate fault effects, which were used to analyze the proposal. We compared the PRSSH approach with a previous proposal, named incremental hypothesis update (IHU), and the MHDIM-HEU with a random heuristic and an optimal branch and bound solution (only for very small network size). Results show the effectiveness of our proposals: PRSSH can decrease the false positive rate up to 66% with respect to IHU, with a significantly reduced processing time (two orders of magnitude). As for MHDIM-HEU, for the same target minimum Hamming distance it requires about half of the symptoms of the random heuristic, with performance very close to the optimal approach. As a third contribution, we evaluated the performance of the PRSSH approach over the codebook optimized by means of MHDIM-HEU, in order to evaluate the impact of the codebook compression on the fault localization performance. Numerical results confirm the goodness of the joint application of both approaches.

Experts have warned that processing of genetic data will soon exceed the computing needs of Twitter and YouTube. This is due to the drop of the costs for sequencing DNA of any living creature and its huge impact in many application areas. Designing suitable network architectures for distributing such data is therefore of paramount importance. Management of genomic data sets is a typical big data problem, characterized not only by a huge volume, but also by the large size of each genomic file. Since it is unthinkable that any professional who needs to process genomes can own the infrastructure for massive genome analysis, a cloud-based access to genomic services is envisaged. This will have a significant impact on the underlying networks, which could become the system bottleneck. In this paper, we propose Genome Centric Networking (GCN), a novel network function virtualization framework for cloud-based genomic data management, designed with the aim of limiting the exchanged traffic by using distributed caching. The key element of GCN is a novel signaling protocol, which allows both discovering network resources and managing caches. We evaluated GCN on a real testbed. GCN allows halving the exchanged traffic and reducing the transfer time of genomic datasets significantly.

The management of network infrastructures has become increasingly complex over time, which is mainly attributed to the introduction of new functionality to support emerging services and applications. To address this important issue, research efforts in the last few years focused on developing software-defined networking solutions. While initial work proposed centralized architectures, their scalability limitations have led researchers to investigate a distributed control plane. Controller placement algorithms and mechanisms for building a logically centralized network view are some examples of challenges addressed in this context. A critical issue that requires specific attention concerns the communication between distributed entities involved in decision-making processes. To this end, we propose SigMA, a signaling framework that supports communication between the different entities of a decentralized management and control system. We also define the communication primitives and interfaces involved in such a decentralized environment. The benefits of SigMA are illustrated through three realistic network resource management use cases with different communication requirements. Based on simulation, we demonstrate the flexibility and extensibility of our solution in satisfying these requirements, thus effectively supporting advanced decentralized decision-making processes.

The management of network infrastructures has become increasingly complex over time, which is mainly attributed to the introduction of new functionality to support emerging services and applications. To address this important issue, research efforts in the last few years focused on developing Software-Defined Networking solutions. While initial work proposed centralized architectures, their scalability limitations have led researchers to investigate a distributed control plane, with controller placement algorithms and mechanisms for building a logically centralized network view, being examples of challenges addressed. A critical issue that has not been adequately addressed concerns the communication between distributed decision-making entities to ensure configuration consistency. To this end, this paper proposes a signaling framework that can allow the exchange of information in distributed management and control scenarios. The benefits of the proposed framework are illustrated through a realistic network resource management use case. Based on simulation, we demonstrate the flexibility and extensibility of our solution in meeting the requirements of distributed decision-making processes.

This paper shows the methodology used for analyzing the technical and non-technical skills necessary for defining a university program providing a Data Science certification. The original aspects of the approach consists of the close collaboration between the organizing universities and the collaborating companies. For what concerns the technical skills, the statistics collected by the already existing programs have been validated by processing thousands of job offers worldwide. The humanistic perspective has completed the picture of the set of skills that have led to the program shown.

In this paper, we present a communication protocol between a pair of biological nanomachines, i.e., a transmitter and a receiver, built upon molecular communications in an aqueous environment. In our proposal, the receiver, acting as a control node, sends a connection setup signal to the transmitter, which stokes molecules, to start molecule transmission. The molecules transmitted by the transmitter propagate in the environment and are absorbed by the receiver through its receptors. When the receiver absorbs the desired quantity of molecules, it releases a tear-down signal to notify the transmitter to stop the transmission. The proposed protocol implements a bidirectional communication by using a number of techniques originally designed for the TCP. In fact, the proposed protocol is connection-oriented and uses the TCP-like probing to find a suitable transmission rate between the transmitter and the receiver to avoid receiver congestion. Unlike the TCP, however, explicit acknowledgments are not used since they would degrade the communication throughput due to the large delay, which is a characteristic feature of molecular communications. Thus, the proposed protocol uses implicit acknowledgments, and feedback signals are sent by the receiver to throttle the transmission rate at the transmitter, i.e., explicit negative feedback. We also present the results of an extensive simulation campaign, used to validate the proposed protocol and to properly dimension the main protocol parameters.

In this paper we analyze the probability of consistency of sensor data distribution systems (SDDS), and determine suitable evaluation models. This problem is typically difficult, since a reliable model taking into account all parameters and processes which affect the system consistency is unavoidably very complex. The simplest candidate approach consists of modeling the state sojourn time, or holding time, as memoryless, and resorting to the well known solutions of Markovian processes. Nevertheless, it may happen that this approach does not fit with some working conditions. In particular, the correct modeling of the SDDS dynamics requires the introduction of a number of parameters, such as the packet transfer time or the packet loss probability, the value of which may determine the suitability of unsuitability of the Markovian model. Candidate alternative solutions include the Erlang phase-type approximation of nearly constant state holding time and a more refined model to account for overlapping events in semi-Markov processes.

In this paper, we evaluate the suitability of the vehicular devices for supporting data storage and distribution applications. Vehicular storage systems have recently emerged as means for making the information related to the vehicular environment available to vehicular users. Information could either be collected by sensors mounted onboard vehicles or coming from the surrounding environment. Modeling a vehicular data storage and distribution system requires the introduction of a number of parameters, such as the vehicle concentration and transmission range, the value of which may determine the suitability or unsuitability of the vehicular environment for establishing a communication infrastructure with the desired quality. In this paper we show how to solve this problem by resorting to a Markovian model and some results achieved through its application.

This paper shows an innovative solution for distributing dynamic sensor data by using distributed caches. Our proposal is based on the concepts of service modularization and virtualization of network nodes made available by the NetServ hosting environment, which has been defined and implemented with the aim of extending the functions of the network nodes. Through a lab experiment involving tens of nodes, we have demonstrated a significant performance improvements in term of traffic saving and download time in comparison with a legacy, Internet-based, approach. Beyond this performance improvements, the proposed solution holds also functional improvements, in terms of dynamic deployment and easy integration with services making use of sensor data.

In this paper, we propose a new gossip-based signaling dissemination method for the Next Steps in Signaling protocol family. In more detail, we propose to extend the General Internet Signaling Transport (GIST) protocol, so as to leverage these new dissemination capabilities from all NSIS Signaling Layer Protocol applications using its transport capabilities. The extension consists of two main procedures: a bootstrap procedure, during which new GIST-enabled nodes discover each other, and a dissemination procedure, which is used to effectively disseminate signaling messages within an Autonomous System. To this aim, we defined three dissemination models, bubble, balloon, and hose, so as to fulfill requirements of different network and service management scenarios. An experimental campaign carried out on GENI shows the effectiveness of the proposed solution.

Autonomic management capabilities of the Future Internet can be provided through a recently proposed service architecture called NetServ. It consists of the interconnection of programmable nodes which enable dynamic deployment and execution of network and application services. This paper shows how this architecture can be further improved by introducing the OpenFlow architecture and implementing the OpenFlow controller as a NetServ service, thus improving both the NetServ management performance and its flexibility. These achievements are demonstrated experimentally on the GENI environment, showing the platform self-protecting capabilities in case of a SIP DoS attack.

In emergency scenarios, like those following a earthquake, road accident in tunnel, large fire and so on, the possibility to track personnel engaged in providing first aid and rescue services is of paramount importance, also in order to promptly manage unexpected and potentially hazardous situations. While in open areas the precision, accuracy and availability of GPS service is enough in most cases, in partially obstructed environments, such as urban canyon, or indoor, GPS service is not available or the quality of provided estimates could be quite coarse. While the literature reports a lot of solutions able to overcome these limitations of the GPS system in standard settings, in emergency scenarios these solutions could be partially or completely unavailable, raising serious concerns about their suitability in these situations. In this paper, we propose a localization and tracking system based on the usage of IEEE 802.11 technology for augmenting the GPS coverage where it is scarce, and to completely substitute it where it is missing, e.g. indoor. The system is characterized by ease of deployment and usage by non-technical personnel, such as that engaged in rescue services in emergency scenarios. In order to assess the feasibility of the proposed solution, we implemented a proof-of-concepts prototype made up of low-cost commercial off-the-shelf components. The preliminary performance tests we performed confirm the effectiveness of proposed system.

Within the activities of the Java APIs for Integrated Networks (JAIN), the Java community offers a set of standard frameworks and open protocol APIs to create advanced telecommunications services. However, recent works have pointed out that Java-based implementations of the Session Initiation Protocol (SIP) stack, the fundamental signalling protocol in the convergent telephone-IP world, perform poorly if they are executed by large multi-core servers, typically used in data centres. The problem lies in the combination of the SIP protocol semantics with the Java language features, which does not allow fully exploiting the computing capabilities of multi-core architectures. To face this problem, we propose a solution to improve the throughput and signalling latency of applications implemented by using Java-based SIP application servers. It consists of the joint usage of virtualization and parallelization techniques. We have performed an extensive measurement campaign by using an open source application server compliant with the JSLEE (JAIN Service Logic Execution Environment) specifications. The rationale of this choice is that JSLEE application servers are currently regarded as very promising candidates for deploying telecom services. Results show that it is possible improving performance in terms of throughput and signalling latency by running more instances of the JSLEE server in parallel, each of them into separate virtual machines deployed on the same server. This improvement can increase throughput values of about 64% and, in the maximum throughput condition, the call set up latency can be nearly halved.

The standard framework proposed by the JAIN SLEE specifications is emerging as a viable alternative for the development of advanced telecom services. The Mobicents JSLEE platform is the only open source implementation of those specifications. Its tuning is not trivial since it relies on several Java technologies implemented in multiple abstract layers. In this paper, we present the implementation of two realistic benchmark services and the relevant test campaign for the performance analysis of the Mobicents JAIN SLEE platform, with the aim of optimizing the achievable system throughput, for using it in a carrier-grade environment.

The WiMAX technology has all the necessary features to effectively provide multicast/broadcast services: (i) it is a broadband access technology, (ii) being wireless, it has intrinsic broadcast capabilities, and (iii) it can provide differentiated quality of service levels. Leveraging on these features, we propose a signaling architecture for multicast/broadcast services exploiting the MBS specification in the Mobile WiMAX standard. The signalling is orchestrated by a JAIN SLEE server. We illustrate this architecture, which is highly modular and based on standard communication protocols (i.e. HTTP, SIP, IGMP), and focus on the design and implementation of the signaling part for a multicast video streaming service.

The growing interest in developing value-added services based on Internet telephony brings into sharp focus several issues related to service creation. It becomes compelling for developers to take advantage of a platform that could offer high-level APIs and standard interfaces, in order to reduce the time-to-market and extend the service portability over multiple networks and devices. The Java community is tackling those issues within the Java APIs for integrated networks (JAIN) activity, utilizing Java enterprise edition technologies in order to ease the development process of telecommunication services. This paper describes a SIP-based call control service for Internet telephony, designed for an IP-PSTN converged scenario. The service has a back-to-back user agent (B2BUA) architecture and is deployed into a service logic execution environment (SLEE). Our SLEE is based on Mobicents, an open-source platform for telecom applications, which offers an implementation of the JAIN SLEE specifications. We present the design and a comparative performance evaluation of different architectural implementations of the service, carried out with SIPp, an open-source SIP traffic generation tool.

In this paper, we present a solution to control automatically a wireless access network and we focus on load balancing target. The reference scenario is that of a single administrative domain that offers a heterogeneous wireless access. We propose a monitoring procedure able to track the access network context, i.e., the current state of resources, the amount of service demand, and the wireless access points available to mobile terminals. To this end, we exploit a cooperative brokerage framework, distributed between terminals and network. The access network context is the input of a selection process that aims to distribute suitably mobile users among available access points. The objective is to balance the users demand, thus improving the quality of the service. The numerical analysis shows the effectiveness of the proposed system in an 802.11b access network.

An improved IP network service (e.g., for real time services) is expected in the near future in both wired and wireless environments. In this regard, handover capabilities are extremely important and challenging, in particular if their use in operation must be seamless. One of the main steps to achieve seamless handover is the quick discovery of the IP addresses and service capabilities of candidate access routers to hand over to. We present a push-mode-multicast based solution to discover and update timely information about wireless resources. We evaluate the effectiveness of the proposed approach in terms of signaling burden and discovery time with respect to solutions already presented in the literature.

In this paper, we consider a per-call, usage-based tariff model to charge for IP services with guaranteed quality. This model is based on the virtual delay, which is a quality of service (QoS) index that describes an improved IP service provided by a network domain. We show how to compute the virtual delay, and how to make it dependent on the service demand. Then, we demonstrate the effectiveness of our tariff model to tune revenues, blocking probability, and resource utilization in a meaningful application scenario. Our goal is to give some directions for network resource dimensioning and pricing purposes, which depend on the service demand.

Traffic engineering and admission control schemes for QoS support over the Internet share the hypothesis that offered traffic must be policed. Recently, measurement-based techniques have been proposed as effective QoS-guarantee mechanisms. For them, a key element is a thorough understanding of the characteristics of regulated traffic. In this paper, we present an analysis of the traffic output process for a leaky-bucket-regulated voice source with silence suppression, which would be a prime candidate for VoIP service. Specifically, we present a procedure for calculating the mean and the variance of the amount of regulated traffic as a function of the measurement period. We evaluate the quality of our results via simulation.

Advanced mobility management is a key issue in providing nomadic users with QoS-enabled services in the future Internet. To improve the mobile IP performance, a crucial role is played by the discovery and selection of the access router to handover. This procedure is of great importance to enable the other additional mobility-related protocols to speed up handover. This paper presents and evaluates a novel approach for carrying out this task, based on a fully distributed process, which also allows timely inter-technology handovers.

In this paper, we present a per-call usage-based tariff model to charge for IP guaranteed services. The tariff is built on the basis of the virtual delay, which is a value characterizing an improved IP service. We propose a model to compute the virtual delay within an administrative domain from purely technical considerations, taking into account traffic, performance and system parameters. In addition, we analyze the sensitivity of the virtual delay and of the tariff to the involved parameters, taking into account both users' benefit and operators' income. Our goal is to provide directions for dimensioning purposes.

IETF developed two main approaches to provide QoS aware services in the Internet: integrated services (IntServ) and differentiated services (DiffServ). Both have well known pros and cons. The stateful IntServ has a greater level of accuracy and a finer level of granularity. The stateless DiffServ possesses excellent scaling properties, but lacks a standardized admission control scheme and, upon overload in a given service class, degradation of service can occur. To provide QoS in DiffServ, three possible strategies are: (i) plain and heavy over-dimensioning; (ii) admission control at the borders of the DiffServ region, coupled with suitable assumptions on the distribution of the traffic within the region, which can lead to over-dimensioning, even if less severe than the previous one; (iii) per-node (i.e., within the region) admission control. Following RFC2990, we proposed an "admission control function which can determine whether to admit a service differentiated flow along the nominated network path", i.e., the third of the above strategies. This function, named GRIP (Gauge & Gate Reservation with Independent Probing), can provide strict QoS guarantees by means of stateless DiffServ-compliant procedures. This feature is paid for with a potential loss of efficiency, with respect to an ideal, stateful admission control. The goal of this paper is to evaluate analytically such loss of efficiency, in a specific scenario. In other words, we estimate how much resources we can waste if we go stateless and avoid state maintenance functions. The comparison between stateless and stateful approaches is performed under the constraint of strictly offering the same performance levels, in terms of, e.g., loss probability and delay.

An innovative architecture to access information transport resources under short-term market conditions is presented. Ultra-wideband (UWB) technology in the centralized wireless access domain is believed to minimize the intrinsic performance limitations of the radio link and to provide flexible access in a very wide range of data rates of over 100 Mb/s, even in a strong multipath environment. The decoupling of user, infrastructure, and service provision in an open mobile access network (OMAN) is proposed to overcome current limitations of radio-based networks. A solution for the joint management of quality-of-service and mobility in the UWB domain is proposed, under the assumption of a resource management compliant with the differentiated services framework. Moreover, a model for the "commoditization" of the network service is introduced. The commodity is identified as the transfer of information units between two end points of the network, measured by a function of the so-called "virtual delay," that summarizes the parameters characterizing the performance of the transfer service. A description is given of this model, together with the definition of a pricing law to charge improved Internet protocol-based services. Such concepts are central in the end-to-end service set-up.

Distributed admission control solutions share the idea that no coordination among network routers (ie, explicit signaling) is necessary, when the decision whether to admit or reject a new offered flow is pushed to the edge of the IP network. Proposed solutions differ in the degree of complexity required in internal network routers, and result in a different robustness and effectiveness in controlling the accepted traffic. This paper builds on a recently proposed distributed admission control solution, called GRIP (Gauge and Gate Reservation with Independent Probing), designed to integrate the flexibility and scalability advantages of a fully distributed operation with the performance effectiveness of admission control mechanisms based on traffic measurements. We show that, in the assumption that traffic sources are dual-leaky-bucket shaped, GRIP allows providing deterministic performance guarantees. Tight QoS performances are made possible even in impulsive load conditions (i.e., sudden activation of several flows), thanks to the introduction of a "stack" mechanism in each network node. A thorough performance evaluation of the conservative effects of the stack shows that the throughput reduction brought about by this mechanism is tolerable, and limited to about 15%.

Current satellite systems operate according to circuit switching transfer modes. To improve flexibility and efficiency, several kinds of packet switching systems have been proposed. However, it appears that full packet switches are still considered complex and expensive, when implemented on board the satellites. For the time being, a compromise has been found in satellite networks with dynamic bandwidth allocation capabilities (DBAC). Such systems are based on classical circuit switches, but the DBAC payload allows the changing dynamically of the capacity of each connection, without the need for tearing-down and setting-up again the connection itself. In this paper we consider a DBAC satellite system and define algorithms to allocate the bandwidth so as to provide deterministic and statistical QoS guarantees. The traffic sources are regulated by standard dual leaky buckets (DLB). We define bandwidth-handling policies, design connection admission control rules and evaluate analytically the system performance. As expected, the numerical results show a significant increase of the overall utilisation factor of our system, when compared with a plain circuit-switching solution.