Description:
The efficient numerical simulation of the environmnent of underground power cables is a challenging task. The Finite Element Method (FEM) encounters huge problems when it is used for the simulation of the aforementioned physical systems. The objective of the research (in cooperation with Terna s.p.a.) is then to devise alternative formulations which overcome the FEM limitations. Integral methods are well-suited to this task.

Hide Description

Description:
Big Data analytics is one of the major challenges for today’s computing systems, aiming at extracting valuable knowledge and information from an exponentially-increasing amount of data. The distributed and heterogeneous nature of the data sources, as well as the huge computational power that is required, is pushing designers towards novel computing systems that combine HPC, Cloud, and IoT solutions (for efficient and distributed computation closer to the data) with Artificial Intelligence (AI) algorithms (for knowledge extraction and decision making). This thesis will focus on data management, understanding how to create a distributed memory architecture for moving the computation closer to the data in a way that is totally transparent to the user.

Hide Description

Description:
The object of this work is to apply techniques based on Operator Precedence (OP) languages for: code verification through model-checking; analysis of massive structured data; runtime verification of critical applications.

Hide Description

Description:
Keywords: Green IT, Fog Computing, Machine Learning, Service-Oriented Architectures. Fog computing aims at creating a continuity between the cloud, with ideally unlimited resources and higher performance and storage capabilities, and the edge, characterized by limited capabilities but lower latency. In this context, new opportunities for managing application deployment are raising together with new challenges. Modern applications are supposed to run in a heterogeneous environment where tasks can be executed dynamically in the cloud and/or in the edge according to the context in which the application is running. These two environments are strongly heterogeneous in terms of resources capabilities, performance, energy efficiency, and power sources. Moreover, the quality of the connection between the different nodes can change dynamically, introducing performance issues. This thesis aims at proposing a dynamic workload allocation algorithm in a performance and energy aware mixed cloud-edge environment taking into account performance and energy efficiency of the available resources and providing an automatic mechanism to decide where to execute a given workload according to the current resource and network state.

Hide Description

Description:
Keywords: Application Monitoring, Fog Computing, Machine Learning, Service-Oriented Architectures. Fog Computing is a paradigm that exploits computational resources available locally (edge resources) and remotely (cloud resources) according to the features of the application, the availability of computational resources in the different nodes, and the data the application needs to consume during its execution. Fog computing is very effective for managing smart buildings where a huge amount of information is generated every minute and moving data from the edge where it is produced to the cloud where it is needed is very expensive and might impact the quality of service. This thesis aims at exploiting the concept of fog computing for building an adaptive monitoring system for smart buildings where adaptive decisions about which data to monitor and where to deploy the applications involved in consuming the data are taken with the aim of reducing the volume of data that has to be moved from the edge to the cloud.

Hide Description

Description:
Discrete event systems (DES) are characterized by a discrete set of states, and state changes are driven by the occurrence of (asynchronous) discrete events. This is a very general class of models, which is often encountered in automated manufacturing, communication and transportation networks, control applications, etc. In timed DES the transitions between activities are further endowed with time bounds in order to represent the duration of activities. A timed control task involves reaching specific states (while avoiding others) within prescribed time limits. The design of a controller for timed DES is a complex and challenging task that requires the development of suitable tools for reachability analysis, path optimization, and control policy design. This thesis aims at the development of a Matlab library of such tools for time Petri nets and its testing on some timed DES control benchmarks.

Hide Description