The continuously increasing complexity of communication networks and the diversity unpredictability traffic demand has led to a consensus view that automation management process is inevitable. Currently, network service techniques are mostly manual, requiring human intervention, leading slow response times, high costs, customer dissatisfaction. In this paper we present AutoNet, self-organizing system for core where robustness environmental changes, namely shifts, topology community interest...

In this article we report on applications and extensions of weighted graph theory in the design control communication networks. We model network as a use existing literature to study its behavior. are particularly interested notions betweenness centrality resistance distance context argue that their most general form, problems can be converted either optimal selection weights or paths based present values graph. Motivated by this, propose two-loop architecture for networks provide directions...

Automated management requires decomposing high-level user requests, such as intents, to an abstraction that the system can understand and execute. This is challenging because even a simple intent performing number of ordered steps. And task identifying adapting these steps (as conditions change) decomposition approach cannot be exactly pre-defined beforehand. To tackle challenges support automated execution, we explore few-shot capability Large Language Models (LLMs). We propose pipeline...

The study of robustness and connectivity properties are important in the analysis complex networks. This paper reports on an effort to compare different network topologies according their algebraic connectivity, criticality, average node degree, betweenness. We consider types behavior these various metrics as scale is increased. Based extensive simulations, we suggest some guidelines for design simplification main finding that, betweenness capture a graph. Depending nature problem at hand,...

Network function visualization (NFV) has emerged as a promising paradigm in networking, where the hardware-based middleboxes are replaced with software-based virtualized entities typically running on cloud to provide specific functionalities. By deploying NFV, network services become more adaptive and cost-effective. Many multicast such real-time multimedia streaming intrusion detection require appropriate chaining; however, NFVs placement well traffic routing strategy guarantee that flows...

Network management complexities stem from the multitude of resources, services, and applications that are to be built on top heterogeneous distributed infrastructure. To address these complexities, a vendor-agnostic, logical, abstract view infrastructure is essential. Intent-based networking (IBN) helps complexity by providing set abstractions (e.g., functional, data, infrastructure), but intelligent automatic decomposition an intent into course actions challenging task. In this article, we...

Auto-scalability has become an evident feature for cloud software systems including but not limited to big data and IoT applications. Cloud application providers now are in full control over their applications' microservices macroservices; virtual machines containers can be provisioned or deprovisioned on demand at runtime. Elascale strives adjust both micro/macro resources with respect workload changes the internal state of whole stack. leverages Elasticsearch stack collection, analysis...

Managing complex infrastructures in multi-domain settings is time-consuming and error-prone. Intent-based infrastructure management a means to simplify by allowing users specify intents, i.e., high-level statements natural language, that are automatically realized the system. However, providing intent-based poses number of challenges: 1) intent translation; 2) plan execution parallelization; 3) incompatible cross-domain abstractions. To tackle these challenges, we propose AppleSeed, an...

This paper proposes a robust Criticality-based Clustering Algorithm (CCA) for Vehicular Ad Hoc NETworks (VANETs) based on the concept of network criticality. Network criticality is global metric an undirected graph, that quantifies robustness graph against environmental changes such as topology. In this paper, we localize notion and apply it to control cluster formation in vehicular wireless network. We use localized node together with universal link measure, Link Expiration Time (LET),...

Intelligent Transportation Systems (ITS) play significant role in the management of traffic congestion large cities. However, current ITS platforms are not suitable for real-time control mainly due to facts that 1) contents rich enough provide detailed information transportation network state, 2) there is no sophisticated notification system alert platform about major issues real-time. Unmanned Autonomous Vehicles (UAV) promising candidates content systems cities and enable notifications,...

With the increased availability of real-time data streams in different domains comes opportunity using this to provide valuable insights into performance systems generating such data. In paper, we are proposing an anomaly detection method be applied on road traffic intelligent transportation systems. The proposed scheme is based multi-channel singular spectrum analysis (MSSA), and aims characterize spatio-temporal properties network. By simultaneously analyzing spatial temporal attributes...

Malicious data manipulation reduces the effectiveness of machine learning techniques, which rely on accurate knowledge input data. Motivated by real-world applications in network flow classification, we address problem robust online with delayed feedback presence malicious generators that attempt to gain favorable classification outcome manipulating features. When delay is static, propose algorithms termed ROLC-NC and ROLC-C when are non-clairvoyant clairvoyant, respectively. We then...

End-to-end delay is a critical attribute of quality service (QoS) in application domains such as cloud computing and computer networks. This metric particularly important tandem systems, where the end-to-end provided through chain services. Service-rate control common mechanism for providing QoS guarantees systems. In this paper, we introduce reinforcement learning-based (RL-based) service-rate controller that provides probabilistic upper-bounds on system, while preventing overuse resources....

As the Internet evolves into a global all-service communication infrastructure, key consideration is providing quality of service guarantees over IP with efficient resource utilization in scalable, flexible, and automatic way. In this article we present virtual network (VN) based architecture for scalable deployment management. Particularly considering DiffServ/MPLS III transport supporting multiple VNs, propose dynamic approach bandwidth sharing among VNs. The service-level-agreement-based;...

In the new digital age, pace and volume of growing transportation related data is exceeding our ability to manage analyze it. this position paper, we present a engine, Godzilla, ingest real-time traffic support analytic mining over data. Godzilla multi-cluster approach handle large volumes data, changing workloads varying number users. The originates at multiple sources, consists types. Meanwhile, belong three camps, namely batch processing, interactive queries graph analysis. language...

This paper reports on a probabilistic method for traffic engineering (specifically routing and resource allocation) in backbone networks, where the transport is main service robustness to unexpected changes network parameters required. We analyze using betweenness of nodes (or links). The theoretical results lead definition "criticality" links. Link criticality used as metric model risk taking specific path from source destination node. Different paths will be ranked based their measure,...

This paper reports on an autonomic network management architecture based the concept of "evolution". A methodology is developed which relying ideas from evolutionary science, virtual networks, and networking. We argue that any communication could be modeled as evolved topology survivability performance requirements. The evolution in direction decreasing chance congestion increasing robustness. describe our system detail tie it to theory evolution. evaluate "betweenness centrality" topologies...

This paper looks at the network robustness problem from a new perspective. Inspired by Darwin's survival value, graph-theoretical metric, betweenness, in combination with weight matrix is used to define global quantity, criticality, characterize adaptability of changes conditions. We show that criticality can be interpreted as average cost journey between any two nodes network, or link betweenness sensitivity network. investigate communication networks particular, and order maximize carried...

In Software Defined Networking (SDN), fine-grained control over individual flow can be achieved by installing appropriate forwarding rules in switches and routers. This allows the network to realize a wide variety of functionalities objectives. But at same time, this flexibility versatility come expense (1) huge burden on limited Ternary Content Addressable Memory (TCAM) space, (2) scalability due large number handled controller. To address these limitations, we present Sector, switch...