@article {Longo2015280, title = {Dependability modeling of Software Defined Networking}, journal = {Computer Networks}, volume = {83}, year = {2015}, note = {cited By 6}, pages = {280-296}, publisher = {Elsevier}, abstract = {

Software Defined Networking (SDN) is a new network design paradigm that aims at simplifying the implementation of complex networking infrastructures by separating the forwarding functionalities (data plane) from the network logical control (control plane). Network devices are used only for forwarding, while decisions about where data is sent are taken by a logically centralized yet physically distributed component, i.e., the SDN controller. From a quality of service (QoS) point of view, an SDN controller is a complex system whose operation can be highly dependent on a variety of parameters, e.g., its degree of distribution, the corresponding topology, the number of network devices to control, and so on. Dependability aspects are particularly critical in this context. In this work, we present a new analytical modeling technique that allows us to represent an SDN controller whose components are organized in a hierarchical topology, focusing on reliability and availability aspects and overcoming issues and limitations of Markovian models. In particular, our approach allows to capture changes in the operating conditions (e.g., in the number of managed devices) still allowing to represent the underlying phenomena through generally distributed events. The dependability of a use case on a two-layer hierarchical SDN control plane is investigated through the proposed technique providing numerical results to demonstrate the feasibility of the approach. {\textcopyright} 2015 Elsevier B.V.

}, keywords = {Availability, Complex networks, Controllers, Degree of distributions, Distributed components, Electric network topology, Information dissemination, Markov processes, Networking infrastructure, Non-Markovian, Quality control, Quality of service, Random processes, Reliability, Reliability and availability, Software defined networking (SDN), Software reliability, Software-defined networkings, Stochastic models, Stochastic systems, Topology, Type expansions}, issn = {13891286}, doi = {10.1016/j.comnet.2015.03.018}, url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84946489290\&doi=10.1016\%2fj.comnet.2015.03.018\&partnerID=40\&md5=b4f32c89d2b7b79fefcaf97082764960}, author = {Francesco Longo and Salvatore Distefano and Dario Bruneo and Marco Scarpa} } @article {Ghosh201457, title = {Scalable analytics for IaaS cloud availability}, journal = {IEEE Transactions on Cloud Computing - IEEE Computer Society}, volume = {2}, number = {1}, year = {2014}, note = {cited By 5}, pages = {57-70}, publisher = {Institute of Electrical and Electronics Engineers Inc.}, abstract = {

In a large Infrastructure-as-a-Service (IaaS) cloud, component failures are quite common. Such failures may lead to occasional system downtime and eventual violation of Service Level Agreements (SLAs) on the cloud service availability. The availability analysis of the underlying infrastructure is useful to the service provider to design a system capable of providing a defined SLA, as well as to evaluate the capabilities of an existing one. This paper presents a scalable, stochastic model-driven approach to quantify the availability of a large-scale IaaS cloud, where failures are typically dealt with through migration of physical machines among three pools: hot (running), warm (turned on, but not ready), and cold (turned off). Since monolithic models do not scale for large systems, we use an interacting Markov chain based approach to demonstrate the reduction in the complexity of analysis and the solution time. The three pools are modeled by interacting sub-models. Dependencies among them are resolved using fixed-point iteration, for which existence of a solution is proved. The analytic-numeric solutions obtained from the proposed approach and from the monolithic model are compared. We show that the errors introduced by interacting sub-models are insignificant and that our approach can handle very large size IaaS clouds. The simulative solution is also considered for the proposed model, and solution time of the methods are compared. {\textcopyright} 2014 IEEE.

}, keywords = {Availability, availability analysis, cloud computing, Downtime, Existence of a solutions, Infrastructure as a service (IaaS), Iterative methods, Lakes, Maintenance, Markov processes, Model driven approach, Numeric solutions, Service level agreement (SLAs), simulation, Stochastic models, Stochastic reward nets, Stochastic systems}, issn = {21687161}, doi = {10.1109/TCC.2014.2310737}, url = {http://www.scopus.com/inward/record.url?eid=2-s2.0-84906976356\&partnerID=40\&md5=a63dc67c92ef7620c8a3b33aca08348b}, author = {Rahul Ghosh and Francesco Longo and Flavio Frattini and Stefano Russo and Kishor S. Trivedi} } @article {Distefano20123701, title = {Investigating dynamic reliability and availability through state-space models}, journal = {Computers and Mathematics with Applications - Elsevier Ltd}, volume = {64}, number = {12}, year = {2012}, note = {cited By 7}, pages = {3701-3716}, abstract = {

Quality standards impose increasingly stringent requirements and constraints on quality of service attributes and measures. As a consequence, aspects, phenomena, and behaviors, hitherto approximated or neglected, have to be taken into account in quantitative assessment in order to provide adequate measures satisfying smaller and smaller confidence intervals and tolerances. With specific regards to reliability and availability, this means that interferences and dependencies involving the components of a system can no longer be neglected. Therefore, in order to support such a trend, specific techniques and tools are required to adequately deal with dynamic aspects in reliability and availability assessment. The main goal of this paper is to demonstrate how state-space based techniques can satisfy such a demand. For this purpose some examples of specific dynamic reliability behaviors, such as common cause failure and load sharing, are considered applying state-space based techniques to study the corresponding reliability models. Different repair policies in availability contexts are also explored. Both Markovian and non-Markovian models are studied via phase type expansion and renewal theory in order to adequately represent and evaluate the considered dynamic reliability aspects in case of generally distributed lifetimes and times to repair. {\textcopyright} 2012 Elsevier Ltd. All rights reserved.

}, keywords = {Availability, Common cause failure, Confidence interval, Dynamic aspects, Dynamic reliability, Load sharing, Markov processes, Markov regenerative process, Markovian, Non-Markovian, Quality of service, Quality standard, Quantitative assessments, Reliability, Reliability and availability, Reliability model, Renewal theory, Repair policy, Semi Markov model, State-space, State-space models, Stringent requirement}, issn = {08981221}, doi = {10.1016/j.camwa.2012.02.038}, url = {http://www.scopus.com/inward/record.url?eid=2-s2.0-84870249108\&partnerID=40\&md5=9ecb143c24c7493a16292faa4f50d175}, author = {Salvatore Distefano and Francesco Longo and Kishor S. Trivedi} } @proceedings {Bruneo20128, title = {Software rejuvenation in the cloud}, journal = {Proceedings of the 5th International Conference on Simulation Tools and Techniques (SIMUTools)}, year = {2012}, note = {cited By 0; Conference of 5th International Conference on Simulation Tools and Techniques, SIMUTools 2012 ; Conference Date: 19 March 2012 Through 23 March 2012; Conference Code:110134}, pages = {8-16}, publisher = {ICST}, address = {Desenzano del Garda, Italy, 19-23 March 2012}, abstract = {

In this paper, we investigate how software rejuvenation can be used in a Cloud environment to increase the availability of a virtualized system composed of a single virtual machine monitor (VMM) on top of which a certain number of virtual machines (VMs) can be instantiated. We start from the assumption that the aging of a VMM increases with the number of VMs it is managing, thus characterizing the problem in terms of dynamic reliability. Therefore, by identifying the age of the VMM with its reliability and based on the conservation of reliability principle, we characterize the time to failure of the VMM through continuous phase type distributions. The system availability is thus modeled by an expanded continuous time Markov chain expressed in terms of Kronecker algebra in order to face the state space explosion and to keep memory of the age reached by the VMM in case the number of the hosted VMs change. Time-based rejuvenation is taken into consideration and the optimal timer is evaluated in order to maximize the VMM availability. Copyright {\textcopyright} 2012 ICST.

}, keywords = {Availability, cloud computing, Continuous phase type distributions, Continuous time Markov chain, Continuous time systems, Endocrinology, Java programming language, Markov processes, phase type distributions, Rejuvenation, Reliability, Reliability principles, Software rejuvenation, Virtual machine monitors, Virtual reality, Virtualized environment}, isbn = {9781450315104}, doi = {10.4108/icst.simutools.2012.247772}, url = {http://www.scopus.com/inward/record.url?eid=2-s2.0-84922767370\&partnerID=40\&md5=a03b06eec703c94d55c0f4875f181902}, author = {Dario Bruneo and Francesco Longo and Antonio Puliafito and Marco Scarpa and Salvatore Distefano} }