@conference {8875501, title = {Enabling Container-Based Fog Computing with OpenStack}, booktitle = {2019 International Conference on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom) and IEEE Smart Data (SmartData)}, year = {2019}, month = {July}, pages = {1049-1056}, abstract = {Operating system-level virtualization using containerization technologies have changed the world of application development and software by bringing flexibility, efficiency and new methods for managing and distributing software. Edge/Fog computing complements nowadays powerful centralized approach leveraging datacenters resources with a number of distributed fog nodes with relatively capable resources in order to provide advanced services in proximity to end users and data sources. In fact, this emerging paradigm provides ubiquitous processing abilities through scattered heterogeneous hardware with different energy availability and computational capabilities. This paper aims at presenting an extension of an IoT centric infrastructure Cloud framework, named Stack4Things, towards the edge through an integration with two of the OpenStack subsystems (i.e., Zun and Kuryr) that deal with containers management.}, keywords = {Application development, Cloud, cloud computing, computer centres, container management, container-based fog computing, containerization technologies, containers, datacenter resources, distributed fog nodes, distributing software, Ecosystems, Edge/Fog computing, Internet of Things, IoT, IoT centric infrastructure cloud framework, Kuryr, Neutrons, OpenStack, OpenStack subsystems, operating system-level virtualization, Servers, software management, ubiquitous processing abilities, virtualisation, Virtualization, Zun}, doi = {10.1109/iThings/GreenCom/CPSCom/SmartData.2019.00181}, author = {Zakaria Benomar and F. Longo and G. Merlino and A. Puliafito} } @conference {9066039, title = {A Mininet-Based Emulated Testbed for the I/Ocloud}, booktitle = {2019 15th International Conference on Mobile Ad-Hoc and Sensor Networks (MSN)}, year = {2019}, month = {Dec}, pages = {277-283}, abstract = {Considering the proliferation of smart devices connected to the Internet, typically going under the aegis of Internet of Things (IoT), a trend has arisen to promote the Cloud paradigm as a suitable management system for such a complex environment. In this context, an effort to extend the OpenStack ecosystem to make it able to support the management of the IoT infrastructure has been made by virtue of the I/Ocloud approach, leading up to its reference implementation, the Stack4Things (S4T) middleware. S4T provides a set of suitable capabilities and features to make the (remote) IoT devices able to join an edge-based IaaS/PaaS Cloud. In the interest of enhancing the S4T middleware scalability and explore new capabilities in particular, ones related to Fog and Edge paradigms, it is becoming a must to test new features in practice at a low financial cost and particular constraints for instance, number/type of devices, network conditions, etc. For this purpose, the use of network emulation tools is a practical and suitable approach. In this paper, we present an integration between the S4T middleware and an emulation tool namely Containernet. Through the integration approach, we model network conditions (e.g., latency, bandwidth, packet loss) and devices (in forms of containers) using Containernet, and we manage the devices (i.e., containers) by means of S4T.}, keywords = {aegis, cloud computing, Cloud paradigm, complex environment, Containernet, containers, Edge computing, Edge/Fog computing, Emulation, input-output programs, Internet of Things, IoT, IoT devices, IoT infrastructure, low financial cost, middleware, Mininet, mininet-based, model network conditions, network emulation tools, OpenStack, OpenStack ecosystem, reference implementation, S4T middleware scalability, smart devices, Stack4Things middleware, suitable management system, Task analysis, Tools, virtualisation}, doi = {10.1109/MSN48538.2019.00060}, author = {Zakaria Benomar and D. Bruneo and F. Longo and G. Merlino and A. Puliafito} } @conference {8784023, title = {Towards trustless prediction-as-a-service}, booktitle = {2019 IEEE International Conference on Smart Computing (SMARTCOMP)}, year = {2019}, month = {June}, pages = {317-322}, keywords = {application program interfaces, Blockchain, cloud computing, cloud provider infrastructure, Computational modeling, Cryptography, Deep Learning, deep learning models, deep neural network models, inference model, inference service, infrastructure acquisition, Machine learning, malicious behaviors, neural nets, neural networks, operation costs, peculiar threat models, prediction API provider, prediction API providers, prediction APIs, prediction-as-a-service, Predictive models, Protocols, security of data, self-hosting costs, software-as-a-service business model, Tendermint}, issn = {null}, doi = {10.1109/SMARTCOMP.2019.00068}, author = {G. Santhosh and F. De Vita and D. Bruneo and F. Longo and A. Puliafito} } @conference {8597889, title = {A Deep Reinforcement Learning Approach For Data Migration in Multi-Access Edge Computing}, booktitle = {2018 ITU Kaleidoscope: Machine Learning for a 5G Future (ITU K)}, year = {2018}, month = {Nov}, pages = {1-8}, keywords = {5G, 5G mobile communication, cloud computing, computational resources, data migration, Deep Reinforcement Learning, deep reinforcement learning approach, distributed services, Edge computing, Keras machine learning, Keras machine learning framework, learning (artificial intelligence), Long Term Evolution, LTE, MEC scenarios, Multi-access Edge Computing, multiaccess edge computing, network performance, OMNeT++/SimuLTE simulator, parameter settings, push data, Quality of service, Reinforcement learning, Resource allocation, resource utilization, Servers, SimuLTE}, doi = {10.23919/ITU-WT.2018.8597889}, author = {F. D. Vita and D. Bruneo and A. Puliafito and G. Nardini and A. Virdis and G. Stea} } @conference {8726749, title = {Extending Openstack for Cloud-Based Networking at the Edge}, booktitle = {2018 IEEE International Conference on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom) and IEEE Smart Data (SmartData)}, year = {2018}, month = {July}, pages = {162-169}, abstract = {Cloud-controlled virtual networking at the edge can be considered a critical feature, highly in demand in the IoT infrastructure management domain. Fostering the vision of the Cloud as a suitable control surface for IoT, an integration between the two ecosystems is achievable at a genuinely Infrastructure-as-a-Service level. This work thus describes an integration design between an IoT-centric infrastructure Cloud framework, already capable of limited networking functionality, and Neutron, the networking subsystem belonging to the core services of the OpenStack platform. Design considerations and trade-offs are detailed in the paper.}, keywords = {Bridges, Cloud, cloud computing, Cloud-controlled virtual networking, control surface, critical feature, Edge computing, IaaS, infrastructure management domain, Infrastructure-as-a-Service level, integration design, Internet of Things, IoT, IoT-centric infrastructure Cloud framework, Linux, network topology, network virtualization, networking functionality, networking subsystem, Neutron, Neutrons, OpenStack, OpenStack platform, Topology, Virtualization}, doi = {10.1109/Cybermatics_2018.2018.00058}, author = {Zakaria Benomar and D. Bruneo and S. Distefano and K. Elbaamrani and N. Idboufker and F. Longo and G. Merlino and A. Puliafito} } @article {35, title = {Adding long-term availability, obfuscation, and encryption to multi-cloud storage systems}, journal = {Journal of Network and Computer Applications}, volume = {59}, year = {2016}, month = {01/2016}, pages = {208-218}, abstract = {

Abstract Nowadays, storage services offer a new way for Cloud providers to do business. This new trend is proved by the number of Cloud storage providers that are continuously appearing on the market. By now, using Cloud storage services is becoming a common practice for end-users. However, the current Cloud storage providers do not offer any guarantees regarding long-term availability and privacy. In fact, data stored in the Cloud could be locked-in, lost, or violated in terms of privacy. In this work, we present an innovative system that on one hand allows end-users to simultaneously rely on different Cloud storage providers in a transparent way and on the other hand to enforce long-term availability, obfuscation, and encryption. Our system is highly reliable, in fact, if a provider is not temporarily or permanently available, end-users continue accessing their data in a secure way. In addition, only the end-users have the full control of the overall security of their data and no sensitive information are disclosed to Cloud storage providers. Several experiments allow us to discuss the performance of our system compared against existing solutions.

}, keywords = {cloud computing, Long-term availability, Privacy, Reliability, Storage Clouds}, issn = {1084-8045}, doi = {10.1016/j.jnca.2014.09.021}, url = {http://www.sciencedirect.com/science/article/pii/S1084804514002288}, author = {A. Celesti and M. Fazio and M. Villari and A. Puliafito} } @article {7060517, title = {Cloud4sens: a cloud-based architecture for sensor controlling and monitoring}, journal = {IEEE Communications Magazine}, volume = {53}, number = {3}, year = {2015}, month = {March}, pages = {41-47}, keywords = {cloud clients, cloud computing, cloud framework, cloud service, cloud-based architecture, Cloud4Sens, Communication standards, Computer architecture, computerised instrumentation, Data models, data-centric model, Databases, device-centric model, environmental data, extensible messaging and presence protocol, Geospatial analysis, IEEE standards, open geospatial consortium, sensing resources management, sensor control, sensor monitoring, sensor web enablement, Sensors, software architecture, Standards, SWE specifications, Temperature sensors, virtual devices, virtual sensing infrastructure, Web services, X509 certificates, XMPP}, issn = {0163-6804}, doi = {10.1109/MCOM.2015.7060517}, author = {M. Fazio and A. Puliafito} } @proceedings {27, title = {Automating the Hadoop configuration for easy setup in resilient cloud systems}, journal = {IEEE Symposium on Computers and Communication (ISCC), 2014}, volume = {Workshops}, year = {2014}, month = {June}, pages = {1{\textendash}4}, publisher = {IEEE Computer Society}, address = {Washington DC}, keywords = {cloud computing, Hadoop, HDFS, Resilient Cloud Storage}, doi = {10.1109/ISCC.2014.6912631}, author = {A. Celesti and M. Fazio and A. Puliafito and M. Villari} } @inbook {50, title = {An Integrated System for Advanced Multi-risk Management Based on Cloud for IoT}, booktitle = {Advances onto the Internet of Things}, volume = {260}, number = {Advances in Intelligent Systems and Computing}, year = {2014}, pages = {253{\textendash}269}, publisher = {Springer International Publishing}, organization = {Springer International Publishing}, address = {Cham}, abstract = {

This chapter presents the Cloud computing technology as strategic solution for the deployment of IoT application and solutions. Cloud computing is a new ICT paradigm able to offer products and solutions as services. Thus, it allows the delivery of on-demand virtual resources (e.g., computational resources, storage systems, applications, data centers,...) over the Internet on a pay-for-use basis. Also, the distributed nature of Cloud computing guarantees high availability of resources dynamically adapting their allocation to specific requirements of the system. The research community together with big business companies are focusing their efforts in the adoption of Cloud for a massive interaction with the physical environment. To show a such trend, we present an on-going project, called SIGMA, which exploits Cloud technologies to acquire, integrate and compute heterogeneous data from several sensor networks for controlling and monitoring both environmental and industrial production systems. Specifically, we describe a new Cloud framework at the basis of the whole SIGMA architecture, in order to show benefits in the adoption of a such technology. The framework is compliant with the Sensor Web Enablement standard specifications and makes use of a plug-in platform to integrate heterogeneous sensing infrastructures. It allows to build abstract objects for accessing sensing devices and observations pandering to the Internet of Things needs.

}, keywords = {cloud computing, Risk management}, issn = {978-3-319-03992-3}, doi = {10.1007/978-3-319-03992-3_18}, url = {http://dx.doi.org/10.1007/978-3-319-03992-3_18}, author = {M. Fazio and A. Celesti and A. Puliafito and M. Villari}, editor = {Gaglio, Salvatore, Lo Re, Giuseppe} } @proceedings {39, title = {The Need of a Hybrid Storage Approach for IoT in PaaS Cloud Federation}, journal = {Proceedings of the 28th International Conference on Advanced Information Networking and Applications Workshops}, year = {2014}, month = {May}, pages = {779{\textendash}784}, publisher = {IEEE COMPUTER SOC}, address = {Washington, DC}, abstract = {

Monitoring activities over many different types of sensors are very challenging to support advanced services for Internet of Things (IoT) and its future. However, one of the major issues is the explosion of the amount of heterogeneous information that has to be stored and processed, thus causing the well known Big Data problem. Some Cloud strategies have been investigated to offer IoT-oriented services, but they do not specifically address solutions for Big Data management. In this paper, we present a two-layer architecture based on a hybrid storage system able to support a Platform as a Service (PaaS) federated Cloud scenario. The proposed architecture combines the benefits of both storage approaches. In particular, it allows us on one hand to extend SQL-like legacy systems, and on the other hand to manage Big Data through an XML-like, non-SQL distributed storage system according to a Cloud federation approach.

}, keywords = {Big Data, cloud computing, federation, IoT, monitoring, PaaS, sensor network}, doi = {10.1109/WAINA.2014.162}, author = {M. Fazio and A. Celesti and M. Villari and A. Puliafito} } @proceedings {34, title = {Resource Management in Cloud Federation Using XMPP}, journal = {IEEE 13th International Symposium on Network Computing and Applications (NCA), 2014}, year = {2014}, month = {Aug}, pages = {67{\textendash}70}, publisher = {IEEE Computer Society}, address = {Washington DC}, abstract = {

This paper deals with Cloud federation issues, where Clouds are both providers and clients of virtual resource at the same time. Specifically, we have designed a solution based on a XMPP communication platform, which allows to set up federated environments and to easily manage virtual shared resources. Heterogeneous and dynamic Clouds can interact in near-real time, making available their own resources according to specific agreement policies. We are thinking to implement the proposed solution using the Hadoop framework, a well-known Cloud Map-Reduce middleware designed to offer different types of services. Here, we analyze the basic and useful elements necessary to making up federated clouds. By mean of Cloud federation, two or more Hadoop clusters can be connected in order to increase service scalability, overcoming some technical limitations of the Hadoop framework itself.

}, keywords = {cloud computing, Cloud services, federation, Hadoop, XMPP}, doi = {10.1109/NCA.2014.15}, author = {M. Fazio and A. Celesti and M. Villari and A. Puliafito} } @proceedings {48, title = {Using Virtualization And noVNC to Support Assistive Technology In Cloud Computing}, journal = {Third Symposium on Network Cloud Computing and Applications (NCCA)}, year = {2014}, pages = {125{\textendash}132}, publisher = {IEEE COMPUTER SOC}, address = {Wa}, abstract = {

In this paper, we focus on how the virtualization technology can support the usage of Assistive Technology (AT) in cloud computing. AT provides software tools to people with disabilities. Unfortunately, these pieces of software are not easy to configure when a disabled user needs to occasionally use a shared computer, e.g., in an internet point, in a library, in a campus, and so on. The proposed solution allows users to access remote Virtual Machines (VMs) through an HTML5 web interface. These virtual environments meet the user{\textquoteright}s personal demand: they support both traditional desktop applications and customized AT tools. In this way, people with disabilities can use any networked computer with a web browser to manage their personal working environments. Nevertheless, achieving such a goal in a cloud computing environment is not trivial at all, because existing remote desktop client viewer applications have not been designed to support the interaction with AT software running on VMs. To achieve such an objective, we propose an open source cloud-based system including no VNC as HTML5 remote desktop proxy/client web application, Virtual Network Computing (VNC) as open source technology to access a remote VM, Oracle Virtual Box as hypervisor, and CLEVER as Virtual Infrastructure Manager (VIM) middleware.

}, keywords = {Assistive Technology, cloud computing, disabled user, HTML5, Virtualization, VNC}, issn = {978-0-7695-5168-5}, doi = {10.1109/NCCA.2014.28}, author = {D. Mulfari and A. Celesti and M. Villari and A. Puliafito} } @booklet {47, title = {Data Reliability In Multi-provider Cloud Storage Service With RRNS}, journal = {Advances in Service-Oriented and Cloud Computing}, volume = {393}, year = {2013}, month = {13}, pages = {83{\textendash}93}, publisher = {Springer - Verlag Berlin Heidelberg}, address = {Berlin}, abstract = {Nowadays, more and more Cloud storage providers are appearing on the market. Nevertheless, data availability and confidentiality represent critical issues considering Cloud computing. This paper discusses an approach that on one hand enables customers to use at the same time different Cloud storage providers, and that on the other hand guarantees both data redundancy and obfuscation. According to our approach, files are fragmented and stored in different Cloud storage providers by means of the Redundant Residue Number System (RRNS). Besides providing us data redundancy, RRNS allows us to preserve the data confidentiality by means of an obfuscation-base strategy spreading metadata over different cloud providers. In addition, our approach allows a customer to retrieve his/her files even if a cloud storage provider is not available anymore. Experiments highlight the factors that have to be considered to configure the system according to the customer{\textquoteright}s requirements.}, keywords = {Big Data, cloud computing, Confidentiality, Reliability, Storage}, url = {http://link.springer.com/chapter/10.1007/978-3-642-45364-9_8}, author = {M. Villari and A. Celesti and F. Tusa and A. Puliafito} } @article {51, title = {A Diversified Set Of Security Features For Xmpp Communication Systems Useful In Cloud Computing Federation}, journal = {INTERNATIONAL JOURNAL ON ADVANCES IN SECURITY}, volume = {6}, year = {2013}, pages = {99{\textendash}110}, abstract = {

Nowadays, in the panorama of Cloud Computing, finding a right compromise between interactivity and security is not trivial at all. Currently, most of Cloud providers base their communication systems on the web service technology. The problem is that both Cloud architectures and services have started as simple but they are becoming increasingly complex. Consequently, web services are often inappropriate. Recently, many operators in both academia and industry are evaluating the eXtensible Messaging and Presence Protocol for the implementation of Cloud communication systems. In fact, the XMPP offers many advantages in term of real-time capabilities, efficient data distribution, service discovery, and inter-domain communication compared to web service technologies. Nevertheless, the protocol lacks of native security features. In this paper, we explore such security issues, discussing how they can be mitigated using both SAML/SASL Single Sign-On (SSO) and XEP 0027.

}, keywords = {authentication, cloud computing, data encryption, digital signature, federation, security, sso, XMPP}, url = {http://www.iariajournals.org/security/tocv6n34.html}, author = {A. Celesti and M. Villari and A. Puliafito} } @proceedings {53, title = {DRACO PaaS: A Distributed Resilient Adaptable Cloud Oriented Platform}, journal = {2013 IEEE 27th International Symposium on Parallel and Distributed Processing Workshops and PhD Forum}, year = {2013}, pages = {1490{\textendash}1497}, publisher = {IEEE Computer Society}, address = {Washington, DC, USA}, keywords = {cloud computing, Distributed System, PaaS, Service Composition}, doi = {10.1109/IPDPSW.2013.266}, author = {A. Celesti and N. Peditto and F. Verboso and M. Villari and A. Puliafito} } @proceedings {49, title = {Using Virtualization And Guacamole/vnc To Provide Adaptive User Interfaces To Disabled People In Cloud Computing}, journal = {The 10th IEEE International Conference on Ubiquitous Intelligence and Computing (UIC-2013)}, year = {2013}, month = {20}, pages = {72{\textendash}79}, publisher = {IEEE COMPUTER SOC}, address = {Washington, DC}, abstract = {

Assistive Technology (AT) provides essential computer accessibility for people with disabilities. In this paper, we examine how Cloud computing can provide adaptive user interfaces to people with disabilities in order to enable them accessing AT tools in form of Software as a Service (SaaS). The proposed solution consists in creating adaptive user interfaces deploying AT tools in Virtual Machines (VMs) which users can manage through the Internet. By using an HTML5-based web interface, computer users with disabilities are able to interact with personalized virtual desktops from any networked computer by means of a web browser without having to setup additional software on the physical machine. Nevertheless, using AT software in web-accessed VMs is not so easy as well as in physical Desktops, because several issues have to be overcome. In order to address such issues, we discuss an architecture integrating Guacamole, i.e., an HTML5 remote desktop gateway, Virtual Network Computing (VNC), i.e., a technology to control a remote Desktop, Virtual Box, i.e., a virtual machine monitor (or hypervisor), and CLEVER a Virtual Infrastructure Manager (VIM) middleware.

}, keywords = {Assistive Technology, cloud computing, HTML5, Physiological, Remote desktop, Virtualization infrastructure, Web browser}, issn = {978-1-4799-2481-3}, doi = {10.1109/UIC-ATC.2013.42}, author = {D. Mulfari and A. Celesti and M. Villari and A. Puliafito} } @article {60, title = {Virtual Machine Provisioning Through Satellite Communications In Federated Cloud Environments}, journal = {Future Generation Computer Systems}, volume = {28}, year = {2012}, month = {01/2012}, pages = {85{\textendash}93}, abstract = {

Cloud federation offers plenty of new services and business opportunities. However, many advanced services cannot be implemented in the real Cloud market due to several issues that have not been overcome yet. One of these concerns is the transfer of huge amount of data among federated Clouds. This paper aims to overcome such a limitation proposing an approach based on satellite communications. By comparing performance in data delivery on the Internet and satellite systems, it is evident that satellite technologies are enough ripe to be competitive against systems with a wired infrastructure. Thus, we propose to make use of satellite transmission to implement fast delivery of huge amount of data. Through the discussion of a use case, where a WEB TV company offers a streaming service, we show how to practically apply the proposed strategy in a real scenario, specifying the involvement of Cloud providers, Cloud users, satellite companies and end-user clients.

}, keywords = {cloud computing, Distributed Cloud service, federation, Satellite communications, Service delivery}, doi = {10.1016/j.future.2011.05.021}, url = {http://www.sciencedirect.com/science/article/pii/S0167739X11001038}, author = {A. Celesti and M. Fazio and M. Villari and A. Puliafito} }