Search

The past few years have witnessed many software companies refactoring their monolithic applications into microservices to take advantage of containers in Cloud-native and DevOps-driven environments. Modern Cloud environments achieve high levels of scalability, flexibility, availability and business agility by combining two different kinds of virtualization: virtual machines and containers. The deployment of complex applications in such environments should be done by taking into account the interactions that may arise from the overlapping of these two different levels of virtualization. In this paper, we first describe a preliminary evaluation of a containerized application that has been obtained by refactoring a monolithic application into microservices. The application was deployed in a private Cloud system that relies on Kubernetes, which is a cluster management software for Docker containers that have been deployed on top of OpenStack-managed KVM virtual machines. Subsequently, in order to systematically investigate the possible interactions that can be produced when multiple containers running different kinds of workloads are activated on top of virtual machines, we present and analyze the results obtained by running different benchmarks in a testbed setup. From the analysis of the benchmark results, we derive some guidelines that should be taken into account by container schedulers when container orchestration platforms, such as Kubernetes, are deployed on top of a virtualized IaaS layer.

In the last few years, there has been a remarkable growth in the number of data centers, which represent one of the leading sources of increased business data traffic on the Internet. An effect of the growing scale and the wide use of data centers is the dramatic increase of power consumption, with significant consequences both in terms of environmental and operational costs. Hence, energy awareness has become one of the major design constraints for Cloud infrastructures. In order to face these challenges, a new generation of energy-efficient and ecosustainable network infrastructures is needed. In this work, a novel energy-aware resource orchestration framework for distributed Cloud infrastructures is introduced, in order to manage both network and IT resources in a typical optical backbone. A high-level overview of the system architecture is provided by focusing on the definition of the different layers of the whole infrastructure, and introducing the Path Computation Element, which is the key component of the proposed architecture. The aim is to explain how both network and IT resources can be managed while, at the same time, the overall power consumption is being minimized and QoS requirements are satisfied. Finally, a green migration plan that is obtained by applying Virtual Machine relocation algorithms is discussed, in order to dynamically react to the fluctuating resource requirements of the VMs.

The Internet of Things paradigm originates from the proliferation of intelligent devices that can sense, compute and communicate data streams in a ubiquitous information and communication network. The great amounts of data coming from these devices introduce some challenges related to the storage and processing capabilities of the information. This strengthens the novel paradigm known as Big Data. In such a complex scenario, the Cloud computing is an efficient solution for the managing of sensor data. This paper presents Polluino, a system for monitoring the air pollution via Arduino. Moreover, a Cloud-based platform that manages data coming from air quality sensors is developed.

Cloud computing has enabled elastic and transparent access to distributed services, without investing in new infrastructures. In the last few years, Cloud computing has grown from being a promising business concept to one of the fast growing segments of the IT industry. Despite of all the hype surrounding the Cloud, enterprise customers are still reluctant to deploy their business in the Cloud. Security is one of the major issues which reduces the growth of Cloud computing and complications with data privacy and data protection continue to plague the market. In this paper, we propose a solution for Hybrid Cloud security, focusing on a Virtual Intrusion Detection System (V-IDS). We present a new architecture that considers the basic principles of the Cloud computing, virtualization and GMPLS Control Plane and applies them to the intrusion detection systems, in order to protect Cloud networks characterized by constantly changing of the underlying infrastructure and physical topology. Based on the defined architecture, we have implemented a prototype of Cloud based IDS that validates our thesis. The prototype is realized though the integration of two open-source technologies: OpenStack and DRAGON (Dynamic Resource Allocation via GMPLS Optical Networks).

The emerging Network Function Virtualization paradigm (NFV) is a disruptive change that creates new business opportunities for vendors and carriers. By exploiting virtualization technologies, it allows the dynamic creation of logically isolated infrastructures over abstracted physical networks. To be able to offer Virtualized Services efficiently, service providers need a new class of Management systems able to manage and orchestrate both network and IT virtualized resources. In this paper, we propose an enhanced network Control Plane, based on the standard Path Computation Element (PCE) architecture, which is able to jointly make network routing and IT server provisioning decisions and that can be used in an NFV environment.

Energy efficiency has become one of the major design constraints for current and future cloud infrastructures. One way to conserve energy is to transition idle clusters, network elements and servers into low-powerstates. Therefore, Virtual Machine placement and dynamic re-configuration algorithms have beenrecently proposed for increasing the energy efficiency of virtualized infrastructures. However, these algorithms usually do not consider the amount of energy consumed by the network infrastructure. In this paper, we present a novel energy-aware resource orchestration framework for distributed cloud infrastructures that manages both Network and IT resources with the goal of minimizing the overall power consumption. We also present the algorithms to be executed in a distributed PCE-compliant system to control VM placement in our environment.

Cloud Computing has risen a great interest over the last years as it represents an enabling technology for flexible and ubiquitous access over the network to a set of shared computing resources. This work comes from an industrial experience aiming at exploiting the cloud potential for virtualizing complex infrastructures such as an entire Air Traffic Center (ATC), a clear example of complex SoS (System of Systems). The use of virtualization is convenient because industry can leverage in-house test-beds to perform distributed testing campaigns in pre-operational phases or to design new automatic fail-over mechanisms for fully distributed systems. In order to realize such mitigation and recovery techniques in the ATC field, indeed, a cloud platform is required to guarantee a low VM provisioning time with the objective of minimizing the service disruption. In this perspective, after having introduced the principal concepts and factors of the provisioning time, we propose a deep analysis and comparison for two different Infrastructure-as-a-Service (IaaS) platforms, namely Open Stack and Open Nebula (using KVM as hypervisor), that were selected through a preliminary scouting phase.

Recent contributions on energy-aware management, aimed at improving the energy efficiency of network infrastructures are based on a combined use of two levers: power consumption optimization obtained by (wholly or partially) switching off a single network device and traffic engineering strategies (with related algorithms) that, given the traffic model on a daily time period, minimize the global energy consumption avoiding traffic congestion. Switches, routers and links of the actual Internet are often represented by simple power consumption models, based on just a few states (typically, on and off). Time for off-to-on switching is often ignored. We claim that complex large-scale network infrastructures need an energy-conscious control plane, which is able of quickly configuring the network by taking into account both energy efficiency goals and QoS constraints. In this paper we present a green extension to the GMPLS network control plane, that is able of minimizing the global power consumption under QoS constraints. Time for fully activating components that were initially in stand-by or off is considered as a QoS requirement.

Follow-Me Cloud (FMC) allows transparent migration of end-points (users and services) in TCP/IP networks. In this paper we describe in details the FMC handover procedures and analyze the parameters that may affect their performance. Our analysis suggests possible optimizations for such procedures.

Software Defined Networking (SDN in short) is reshaping the future of computer networks. By decoupling control and data planes, SDN technologies allow a more flexible management of network infrastructures, whose resources may be operated by means of a well defined programming interface. Several approaches have been recently proposed to implement the SDN concept. OpenFlow is maybe the most prominent SDN component, having been supported by several device vendors. This paper discusses a practical experience in designing an OpenFlow controller for a Mobile Cloud Management system. We present the programming model and the designed abstraction and discuss the lesson learned.

Cloud-enabled datacenters need advanced support for an integrated management of platform virtualization technologies. The networking infrastructure of large-scale datacenters is implemented according to redundant multi-tiered architectures whose layers operate at Layer 3 of the networking stack. Splitting the network infrastructure of a datacenter in a number of IP subnets, however, creates limits to the migration of Virtual Machines, reducing the possibility for administrators to efficiently balance the load and reduce the energy consumption of the whole infrastructure. In this paper we propose an innovative solution that allows transparent migration of Virtual Machines across the whole datacenter, based on the coordinated use of NAT rules that need to be consistently managed across the layers of the datacenter networking infrastructure. We describe in details how our approach can be easily implemented with current network devices without any modification to their hardware and present an experimental evaluation of an early prototype of our solution.

Wireless Mesh Networks (WMNs) have recently gained wide attention as a way to provide Internet connectivity in hard-to-reach areas as well as in big cities. The main problem of such networks is their limited bandwidth availability compared to wired networks. Hence, efficient usage of available communication resources is of crucial importance for WMNs. Peer-to-peer applications, these days, produce a considerable amount of traffic over the Internet. Such applications build their own overlay network by not taking into account the topology and the status of the underlying infrastructure, so causing an inefficient utilization of physical communication resources. This problem is particularly severe for Wireless Mesh Networks. In this paper we propose a cross-layer solution for P2P Bittorrent traffic optimization in WMNs, based on the ALTO (Application Layer Traffic Optimization) Service recently defined by the IETF. Our solution is able to significantly optimize Bittorrent traffic and it does not require any modification to the Bittorrent application. The tests we performed show a reduction of about 27% of download time and a reduction of about 66% of traffic crossing the WMN borders.

The ultimate success of the Wireless Mesh Network paradigm (WMN) in large scale deployments depends on the ability to test it in real world scenarios. A typical application scenario which is worth to be investigated in such a context is peer-to-peer traffic management. The creation of large scale testbeds for evaluating wireless mesh technologies and protocols, and for testing their ability to support real world applications in realistic environments, is then a crucial step. OMF (cOntrol and Management Framework) is a well-established control, measurement, and management framework for wireless testbeds. In this paper we present how we integrated an OMF-based wireless testbed in the planetary-scale PlanetLab testbed, making it possible for PlanetLab users to run experiments spanning on both PlanetLab and an OMF-based wireless testbed. In order to demonstrate the usefulness of such an integrated scenario, we tested on it an innovative peer-to-peer traffic optimization technique for the BitTorrent file sharing application. The possibility of running this kind of experiments highlighted several real-world issues which could be investigated thanks to our hybrid experimental scenario.

The success of the Cloud Computing paradigm may be jeopardized by concerns about the risk of misuse of this model aimed at conducting illegal activities. In this paper we address the issue of detecting Denial of Service attacks performed by means of resources acquired on-demand on a Cloud Computing platform. To this purpose, we propose to investigate the consequences of the use of a distributed strategy to detect and block attacks, or other malicious activities, originated by misbehaving customers of a Cloud Computing provider. In order to check the viability of our approach, we also evaluate the impact on performance of our proposed solution. This paper presents the installation and deployment experience of a distributed defence strategy and illustrates the preliminary results of the performance evaluation.

In many areas of ICT the use of virtualization techniques has decoupled the binding between physical resources and functional roles assigned to them. While the use of virtualization in computing systems is already extremely popular, similar forms of virtualization have been proposed for the network infrastructure as well. In such a context, we propose a new paradigm that aims at extending the concept of virtualization to network services, by decoupling service execution environments and their physical location. We call this paradigm Service Switching. In a Service Switching environment, service instances may be dynamically migrated across geographically dispersed data centers to pursue better usage of both network and computing resources.