Routers use queues to temporarily hold packets when they can’t process them immediately. Congestion occurs when the arrival rate of packets exceeds the processing capacity, leading to increased queuing delay. Active Queue Management (AQM) strategies aim to reduce congestion and queuing delay by draining packets from queues. Traditionally, AQMs are placed in the Egress pipeline of Programmable Data Plane (PDP) hardware due to the availability of queue delay information there. We argue that this approach wastes the router’s resources because the dropped packet has already consumed the entire pipeline of the device. We propose ingress Random Early Detection (iRED) as a more efficient P4-AQM approach fully implemented in PDP hardware, that introduces the concept of disaggregated AQM, effectively resolving the egress drop problem. iRED also supports the Low Latency, Low Loss, and Scalable Throughput (L4S) framework, saving device pipeline resources by dropping packets in the Ingress block. Our experiments with a Tofino2 programmable switch achieved a rate of 1Tbps and show that iRED can reduce router resource consumption by up to 10x in memory usage, 12x fewer processing cycles, and 8x less power consumption for the same traffic load. It achieves fairness in bandwidth usage for different types of traffic and significantly improves Quality of Service (QoS) in Dynamic Adaptive Streaming over HTTP (DASH) with up to a 2.34x improvement in Frames Per Second (FPS) and a 4.77x increase in video player buffer fill. We believe this is a pioneering work to highlight the egress drop problem and deeply explore its potential damage to network performance.

The development of techniques able to forecast the mobile network traffic in a city can feed data driven applications, as Virtual Network Functions (VNF) orchestrators, optimizing the resource allocation and increasing the capacity of mobile networks. Despite the fact that several studies have addressed this problem, many did not consider neither the traffic relationship among city regions nor the information retrieved from public transport stations, which may provide useful information to better anticipate the network traffic. In this paper, we propose a new deep learning based architecture to forecast the network traffic using representation learning and recurrent neural networks. The framework, named Mobile Traffic Predictor Enhanced by Neighboring and Transportation Data (MTP-NT), has two major components: the first one is responsible of learning from the time series of the region to be predicted, with the second one learning from the time series of both neighboring regions and public transportation stations. Several experiments were conducted over a dataset from the city of Milan, as well as comparisons against widely adopted and state-of-the-art techniques. The results shown in this paper demonstrate that the usage of public transport information contributes to improve the forecasts in central areas of the city, as well as in regions with aperiodic demands, such as tourist regions.

Network slicing (NS) is becoming an essential element of service management and orchestration in communication networks, starting from mobile cellular networks and extending to a global initiative. NS can reshape the deployment and operation of traditional services, support the introduction of new ones, vastly advance how resource allocation performs in networks, and notably change the user experience. Most of these promises still need to reach the real world, but they have already demonstrated their capabilities in many experimental infrastructures. However, complexity, scale, and dynamism are pressuring for a Machine Learning (ML)-enabled NS approach in which autonomy and efficiency are critical features. This trend is relatively new but growing fast and attracting much attention. This article surveys Artificial Intelligence-enabled NS and its potential use in current and future infrastructures. We have covered state-of-the-art ML-enabled NS for all network segments and organized the literature according to the phases of the NS life cycle. We also discuss challenges and opportunities in research on this topic.

Video services account for the largest share of all Internet traffic, demanding a network capable of supporting the requirements of delay-sensitive traffic. Fluctuations in network load can cause high delays in the queues of network routers, which tend to degrade the Quality of Service (QoS) for adaptive video streaming, such as Dynamic Adaptive Streaming over HTTP (DASH). This work is positioned in the scope of active management queues (AQM) to improve the QoS of a DASH service by means of dropping packets. One traditional AQM that adopts a packet drop policy is Random Early Detection (RED), developed to drain the flow in times of congestion and thus reduce queueing delay. We revisited and implemented a P4-based implementation of RED, named iRED (ingress RED), an algorithm capable of dropping packets at the ingress pipeline, an innovation compared to other AQM strategies based on dropping at the egress. iRED was evaluated in two scenarios. First, we compare iRED against state-of-art AQM algorithms employing egress packet dropping in terms of Round-Trip Time (RTT), throughput and their impact on resources usage. Our findings indicate that iRED outperforms existing P4-based approaches by approximately up to 2.5x in RTT and 0.75x in throughput for the given buffer sizes. Next, we compare iRED versus Tail Drop (TD) approach in an emulated programmable Content Delivery Network (CDN) employing DASH. Experiments indicate that the iRED improve the QoS by approximately 0.85x in terms of cached video available in the client’s buffer and 0.9x in Frames Per Second (FPS) played.

Network softwarization is one of the key enablers of the future Internet evolution, also supporting the road from the fifth generation (5G) to the next-generation communication systems, namely 6G, with their main objective of bringing hyper-connected experience to every corner of society.

The Softwarization of networks is enabled by the SDN (Software-Defined Networking), NV (Network Virtualization), and NFV (Network Function Virtualization) paradigms, and offers many advantages for network operators, service providers and data-center providers. Given the strong interest in both industry and academia in the softwarization of telecommunication networks and cloud computing infrastructures, a series of special issues was established in IEEE Transactions on Network and Service Management, which aims at the timely publication of recent innovative research results on the management of softwarized networks.

A key feature of the cloud network slicing concept is the dynamic (de)provision of end-to-end infrastructures composed by computing, network, and storage resources, in order to meet the service needs of a variety of vertical industries. The resulting resource ensemble needs to be instantiated over a (potentially large) number of administrative and technological domains – a complex challenge for the management and orchestration of the allocated resources. Resource monitoring in this new so-called slice entity is extremely important in order to address the above operations. Therefore, in this work, we discuss the design and implementation of an elastic architecture for the monitoring of physical and virtual resources in cloud network slices that span both multiple administrative and technological domains. Since the concept of cloud network slices is quite recent in the literature, we highlight that our results present the performance evaluation of the proposed monitoring architecture when instantiating, stopping or performing elasticity operations on the slices. Additionally, we also analyze the performance of the architecture by scaling the number of metrics being monitored in overloaded scenarios.

The NECOS (Novel Enablers for Cloud Slicing) platform controls the entire lifecycle of end-to-end cloud-network slice instances, so that new services and applications can run in a broader perspective for quality-guaranteed and highly isolated multi-tenancy. Elasticity plays a vital role in ensuring the efficiency of the NECOS platform. According the state of the art, mainstream reactive elasticity solutions are not suitable for dealing with cloud-network slicing-defined systems by the fact that they cannot trigger time-burdensome vertical elasticity on deriving resource-depleting conditions. This gap can be filled by introducing in the elaSticity cLOud-neTwork Slices (SLOTS) solution, which participates in the NECOS platform in the form of new building blocks that interwork with others to add new patterns of cloud- and networking-type resources under resource-depleting conditions. The unique contributions made by SLOTS are as follow, and go beyond the state–of-the-art by: (i) fully supporting cloud-network slice-defined systems; (ii) offering a hybrid elasticity approach; (iii) provide a sophisticated mechanism to enable a slice solidarity approach, and (iv) providing a real lab-premised large-scale testbed for a SLOTS prototyping assessment. Analysis on the evaluation results suggest that SLOTS performs well by accomplishing a higher number of elasticity events under resource-critical conditions, while slightly adding computing overhead, when compared with a widely-used stochastic-based representing solution.

To support future 5G computing and communication scenarios, cloud-network management tools should deploy cloud-network services adopting uncomplicated ways, reducing not only the time to market but also broadening the community capable of deploying new services. In this paper, we present the support of NECOS Platform, an EU-Brazil jointly funded project, towards slice-as-a-service creation from non-structured service description. We describe how NECOS architecture allows such functionality during the slice creation loop, and we present the initial efforts we took for structuring such a mechanism.

Network Function Virtualization (NFV) and Machine Learning (ML) are envisioned as possible techniques for the realization of a flexible and adaptive 5G network. ML will provide the network with experiential intelligence to forecast, adapt and recover from temporal network fluctuations. On the other hand, NFV will enable the deployment of slice instances meeting specific service requirements. Moreover, a single slice instance may require to be deployed across multiple substrate networks; however, existing works on multi-substrate Virtual Network Embedding fall short on addressing the realistic slice constraints such as delay, location, etc., hence they are not suited for applications transcending multiple domains. In this paper, we address the multi-substrate slicing problem in a coordinated manner, and we propose a Reinforcement Learning (RL) algorithm for partitioning the slice request to the different candidate substrate networks. Moreover, we consider realistic slice constraints such as delay, location, etc. Simulation results show that the RL approach results into a performance comparable to the combinatorial solution, with more than 99% of time saving for the processing of each request.

The softwarization of networks is enabled by the SDN (Software-Defined Networking), NV (Network Virtualization), and NFV (Network Function Virtualization) paradigms, and offers many advantages for network operators, service providers and datacenter providers. Given the strong interest in both industry and academia in the softwarization of telecommunication networks and cloud computing infrastructures, a series of special issues was established in IEEE Transactions on Network and Service Management, which aims at the timely publication of recent innovative research results on management of softwarized networks.

Service assurance for cloud applications is a challenging task and is an active area of research for academia and industry. One promising approach is to utilize machine learning for service quality prediction and fault detection so that suitable mitigation actions can be executed. In our previous work, we have shown how to predict service-level metrics in real-time just from operational data gathered at the server side. This gives the service provider early indications on whether the platform can support the current load demand. This paper provides the logical next step where we extend our work by proposing an automated detection and diagnostic capability for the performance faults manifesting themselves in cloud and datacenter environments. This is a crucial task to maintain the smooth operation of running services and minimizing downtime. We demonstrate the effectiveness of our approach which exploits the interpretative capabilities of Self- Organizing Maps (SOMs) to automatically detect and localize different performance faults for cloud services.

The softwarization of networks is enabled by the SDN (Software-Defined Networking), NV (Network Virtualization), and NFV (Network Function Virtualization) paradigms, and offers many advantages for network operators, service providers and datacenter providers. Given the strong interest in both industry and academia in the softwarization of telecommunication networks and cloud computing infrastructures, a series of special section was established in IEEE Transactions on Network and Service Management, which aims at the timely publication of recent innovative research results on management of softwarized networks.

Automated detection and diagnosis of the performance faults in cloud and datacenter environments is a crucial task to maintain smooth operation of different services and minimize downtime. We demonstrate an effective machine learning approach based on detecting metric correlation stability violations (CSV) for automated localization of performance faults for datacenter services running under dynamic load conditions.

Predicting resource requirements for cloud services is critical for dimensioning, anomaly detection and service assurance. We demonstrate a system for real-time estimation of the needed amount of infrastructure resources, such as CPU and memory, for a given service. Statistical learning methods on server statistics and load parameters of the service are used for learning a resource prediction model. The model can be used as a guideline for service deployment and for real-time identification of resource bottlenecks.

We use statistical learning to estimate end-to-end QoS metrics from device statistics, collected from a server cluster and an OpenFlow network. The results from our testbed, which runs a video-on-demand service and a key-value store, demonstrate that the learned models can estimate QoS metrics like frame rate or response time with errors bellow 10% for a given client. Interestingly, we find that service-level QoS metrics seem "encoded" in network statistics and it suffices to collect OpenFlow per port statistics to achieve accurate estimation at small overhead for data collection and model computation.

Softwarization of networks is an important trend, enabled by the NV (Network Virtualization), SDN (Software-Defined Networking), and NFV (Network Function Virtualization) paradigms and offers many advantages for network operators, service providers and datacenter providers. Given the strong interest in both industry and academia in the softwarization of telecommunication networks and cloud computing infrastructures, a series of special issues was established in IEEE Transactions on Network and Service Management, which aims at the timely publication of recent innovative research results on management of softwarized networks.

Brazilian public transportation systems are facing a significant demand reduction, mainly due to the poor quality of the offered services, lack of information regarding lines and timetables, high cost and lack of investment from the government. Even though it is not trivial to improve financial aspects related to the public transportation system, this work claims that the overall system quality can be improved through ubiquitous data collection according to a proposed ontology, which is the basis for knowledge extraction to support the required quality of experience improvements. The proposed architecture relies on standard technologies available nowadays, providing a low cost solution for the required data collection and analysis. This paper presents the proposed inter-networking architecture and ontology, then evaluates the system performance using a prototype developed with standard solutions.

This paper introduces and evaluates a Traffic Sign Management Architecture (TSMA), which represents a paradigm shift for the deployment of traffic sign infrastructure in the context of Intelligent Transport Systems, Vehicular Networks and Smart Cities. The proposal addresses limitations of the current traffic control model by enabling remote updates of traffic signs and displaying them on the vehicular navigation system display to improve their legibility. TSMA is an architecture developed to provide V2I interaction using a commodity technology, Wi-Fi, through the beacon-stuffing technique. The initial design of TSMA's security mechanisms is also presented in this paper. Evaluations were performed on a developed prototype and simulation environments.

In mobile ad hoc networks (MANET) nodes are free to move on the environment and interact with each other and with the infra-structure, enabling a multitude of applications and services. However, the same mobility that is key to MANET is also the greater limiting factor in the quality of the services provided in this environment, since the infrastructure must keep adapting to the mobility. This paper describes algorithms for predicting the future position of mobile nodes in MANET, allowing the infrastructure to proactively adapt. Our algorithms maintain the recent movement history of nodes in a compact representation, a graph, based on stigmergy of ant colonies. Predictions are generated through a limited depth first search. We have experimented this method against real world data and the results show that the prediction is accurate for short time horizons (30 seconds) in a metropolitan area (Seattle) divided in a grid of two or three-block square cells, in 77.8% of the cases. This method may be used in optimizing MANET and enabling novel applications, for example, proactive hand-off, tailored content generation, and proactive routing.

This paper presents a proposal for shrinking the number of IPv4 FIB (Forwarding Information Base) entries required on routers. Traffic forwarding under the proposed mechanism is based on the current ASNs (Autonomous System Numbers), and can be gradually adopted by ISPs.We find that, at the cost of adding 8 bytes per packet, the proposed ASN-FWD technique is capable of providing full IPv4 traffic forwarding based on ASN information, which is correspondent to 10% of the current number of IPv4 prefixes present on FIB of routers. Among its main benefits, the proposed approach alleviates the pressure on the amount of FIB shipped on routers, and paves the way for a worldwide adoption of IPv6.

The information society is facing a sharp increase in the amount of information driven by the plethora of new applications that sprouts all the time. The amount of data now circulating on the Internet is over zettabytes (ZB), resulting in a scenario defined in the literature as Big Data. In order to handle such challenging scenario, the deployed solutions rely not only on massive storage, memory and processing capacity installed in Data Centers (DC) maintained by big players all over the globe, but also on shrewd computational techniques, such as Big Table, MapReduce and Dynamo. In this context, this work presents a DC structure designed to support the similarity search. The proposed solution aims at concentrating similar data on servers physically close within a DC, accelerating the recovery of all data related to searches performed using a primitive get(k, sim), in which k represents the query identifier, i.e., the data used as reference, and sim a similarity level.

The semantic meaning of a content is frequently represented by content vectors in which each dimension represents an attribute of this content, such as, keywords in a text, colors in a picture or profile information in a social network. However, one important challenge in this semantic context is the storage and retrieval of similar contents, such as the search for similar images assisting a medical procedure. Based on it, this paper presents a new Distributed Hash Table (DHT), called Hamming DHT, in which Locality Sensitive Hashing (LSH) functions, specially the Random Hyperplane Hashing (RHH), are used to generate content identifiers, propitiating a scenario in which similar contents are stored in peers nearly located in the indexing space of the proposed DHT. The evaluations of this work simulate profiles in a social network to verify if the proposed DHT is capable of reducing the number of hops required in order to improve the recall in the context of a similarity search.

This paper presents the URBAN XOR1protocol, an XOR-based flat routing mechanism developed for vehicular ad hoc networks (VANETs) formed in urban scenarios. The paper firstly describes the URBAN_XOR routing principle, which requires reduced knowledge about the set of nodes present in the VANET in order to provide traffic forwarding. Basically, the URBAN XOR protocol introduces the concept of local visibility, prioritizing the insertion of closer neighbors (in number of hops) in the routing tables, and simplifying the management of the frequent network mobility found in VANETs. Then, the performance of the proposed URBAN_XOR protocol is compared through simulation with other topology-based and position-based protocols, characterizing its performance in terms of path availability ratio, end-to-end delay, path length and path duration. The results reveal that URBAN_XOR contributes for the overall network stability, reducing the end-to-end delay due to its ability of generating shorter paths. At the same time, URBAN_XOR exhibits path availability similar to other topology-based protocols, but exhibiting better path duration times.

Many online data sets evolve over time as new entries are slowly added and existing entries are deleted or modified. Taking advantage of this, systems for incremental bulk data processing, such as Google's Percolator, can achieve efficient updates. To achieve this efficiency, however, these systems lose compatibility with the simple programming models offered by non-incremental systems, e.g., MapReduce, and more importantly, requires the programmer to implement application-specific dynamic algorithms, ultimately increasing algorithm and code complexity.

In this paper we present a performance analysis of XOR1-based flat routing protocols in high mobility conditions, considering a vehicular ad hoc network (VANET) formed in a highway scenario. First, we describe an XOR-based protocol that incorporates several adaptations of the existing XOR-based routing algorithms for wired networks, in order to cope with the network mobility. Then we propose an improved version of it, XORi, which modifies the protocol's information gathering process to accommodate the specific dynamic nature of VANETs topology. Finally, we evaluate the performance of XOR-based protocols with other topology-based routing protocols. Simulation results allow us to characterize the performance of this class of protocols through the comparison of the packet delivery ratio, end-to-end path delay and average number of path hops2. When a moderate density of nodes is considered, simulations show that XOR-based algorithms achieve almost the same packet delivery rate as link state algorithms, such as OLSR, while for high density of nodes XOR-based algorithms scale better in terms of delay when compared to source routing algorithms, such as DSR3.

A frequent subject in forums, academia and industry is the evolution of the Internet in terms of routing. Basically, the scalability in the Default Free Zone related to (1) the growing rate of the routing tables and (2) the convergence of the routing system, are pointed by routing experts as the main concerns of the current mechanism. Several approaches have emerged, but they normally require a mapping system to translate from identifiers to locators (IP). We contribute with this discussion by introducing our XOR-based Flat Routing mechanism for Internet-like topologies. Essentially, the proposed mechanism routes directly on top of flat ASes identifiers, eliminating the need for mapping systems. In this work we propose a mechanism for building the routing tables over the XOR-based scenario in conjunction with a reachability service developed using the concepts of Landmark and Bloom filters. The proposal is evaluated using our developed emulation tool under five different Internet-like topologies ranging from 512 to 8192 nodes.

Flat routing is subject of many proposals found in the literature. One challenging task in this scenario is the flat identity space management since aggregation is not possible. The Landmark-based Flat Routing (LFR) proposal is an alternative to address the routing tables' growth in a scenario which (1) organizes the network in regions represented by landmarks and (2) adopts an XOR-based routing mechanism inside regions. In this work, Bloom filters are introduced to LFR as an effective mechanism to deal with two main aspects of flat routing. The first one is to advertise aggregated reachability information between landmark regions and the second combines the intra-region XOR-based mechanism with Bloom filters in order to reduce unnecessary walk at the flat binary space, opening up many possibilities for the Flat Routing scenario, especially for stretch reduction and scalability for disseminating ``reachability'' information. The evaluations were done in a regular mesh network topology with 256 nodes and the results show the advantages of having Bloom filters in both scenarios, highlighting the effectiveness of such mechanism for flat identifiers aggregation.

In this paper we make the domain entity a first class citizen. The concept of Domain Identifiers (DIDs) is introduced to effectively bring the domains to the next generation Internet scenario. The paper presents an architecture to address challenging next generation Internet requirements such as node and domain mobility, multi-homing, security, network composition and inter-domain routing. Although our architecture supports all the mentioned requirements, the focus of this paper is specifically on node and domain mobility in order to evaluate and compare the advantages of having DIDs for facilitating the mobility. First, we present the generic Next Generation Internet architecture proposal and its envisioned scenario. Then, we show how to instantiate it to work together with the current Internet. We have developed a prototype to evaluate our proposal and we depict results related to node and domain mobility under this gradual deployment scenario using the Internet as the core.

The current internetworking architecture presents some limitations to naturally support mobility, security and multi- homing. Among the limitations, the IP semantic overload seems to be a primary issue to be considered. In this paper we present a next generation internetworking architecture to overcome the IP semantic overload by introducing an identity layer located between the network and transport layers. This new layer provides a stable cryptographic identifier for end-hosts and seamlessly allows the deployment of new services, such as mobility, multi-homing and security. A prototype was implemented and evaluated considering some mobility scenarios, including intra-domain, inter-domain and simultaneous node mobility.