Here you can find a list of available thesis. I am also to supervise other thesis related to Green Information Systems, Adaptive Information Systems, Fog Computing, Business Process Management, and Process Mining.
Improving Energy Efficiency of Service-based Applications in Fog Computing
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.
KEYWORDS: Green IT, Fog Computing, Machine Learning, Service-Oriented Architectures
An Adaptive Monitoring System for managing Smart Buildings in Fog Computing
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.
KEYWORDS: Application Monitoring, Fog Computing, Machine Learning, Service-Oriented Architectures.
Customized Data as a Service in heterogeneous environment
The Data as a Service paradigm enables the access to the same data source to several users or applications hiding the complexity of retrieving the data providing general interfaces. This service cannot be general, since different users might have different requirements related to both the quality of the data and the quality of the service itself (e.g., response time and availability of the DaaS). This thesis will focus on a customizable DaaS, able to provide a personalised service to each customer according to the different requirements. The DaaS will be in charge of managing the data sources and their interrogation to ensure the quality levels required by the different customers. The DaaS is in charge of managing several data sources and several copies of the same data source, and can enact data movement actions, consisting in moving data form one storage to another. Storages might be both in the edge of the internet or in the cloud, and movement actions might be conditioned by privacy and security constraints on the data sources. Moreover, the DaaS provides several pre-computation and transformations of the data sources to make them ready for the usage of the different customers. Also, this computation (which can be modelled as the composition of several micro-services) can be executed both in edge and cloud computational resources. Goal of the thesis is the design of a distributed decision system, able to take decision on data movement and computation deployment in order to ensure the data quality and quality of service requirements of several customers.
KEYWORDS: DaaS, data quality, distributed decision system, Edge computing.
Energy-aware management of distributed applications in heterogeneous environments
Energy consumption of IT is an issue that has been explored in the current literature from several perspectives, mainly focussing on the data center and cloud management point of view. However, responsibility on energy consumption is also shared with the application owners, which express performance requirements that might be in contrast with the energy efficiency best practise. According to this, it is important to make application owner aware of the energy impact of their applications. Modern applications are not monolithic (running on a single virtual machine) but are executed as a composition of micro-services and functions that are dynamically composed to provide the global service. Each of these components can be executed on resources with different features. As an example, several applications are moving their computation execution towards the edge of the internet, where small data-centers or IoT devices can be employed to host the computation improving the overall performance. In this dynamic and heterogeneous scenario, it is difficult to assign the energy impact to an application. This thesis will explore this issue by designing dynamic application through dynamic business processes, in which activities composing the process can change over time.
KEYWORDS: Green IS, energy-aware applications, Edge computing.
Energy-aware Cloud Systems Management
The amount of services provided by modern companies has increased exponentially in the last years, generating a huge amount of data that needs to be processed and analysed. To improve their efficiently and reduce costs, companies are shifting their services to the Cloud. However, lower costs are increasing the demand for cloud services with a consequent increment in the size and kind of information to be maintained. A question is arising and is becoming a driver for innovation: “Is it sustainable?”. This thesis focuses the attention over the explosion of the Internet of Things ecosystem, that is bringing more devices connected that people in the world, and the Big Data challenge that arises in modern ISs, which have an impact over the energy consumption due to their maintenance and analysis.
KEYWORDS: Green IS, big data, Cloud based Information Systems.
Sensing Environment – smart data collection
The environment in which we live every day is surrounded by a high number of devices and sensors able to capture a big amount of information. Internet of Things (IoT) and Ambient Intelligence (AmI) are just examples of sensing environments producing a big amount of data and/or events. The thesis should investigate the management of the data collected by the sensing environment. Data have to be collected in a distributed way in order to reduce the traffic of data from their collection to the storage. Techniques are going to be investigated for optimizing this process, dynamically adapting the set of data to collect and the sample time according to their relevance. Data relevance is supposed to be dependent on the specific state of the system (e.g. the variability of the information during the observation in a time interval).
KEYWORDS: IoT, efficient data management, big data, distributed monitoring system.
Process evolution discovery applied to ambient assisted living
Information systems have to deal with several processes which models the way the activities of the organization work. However, the organization and its processes are continually evolving and the models of the organization processes could not reflect the reality. This thesis should analyse the information produced during the processes execution in order to discover hidden information about the process execution and behaviour. This analysis will be addressed to compare the process behaviour with the expected one, to detect unexpected behaviours, and to automatically understand if and when the process has changed and the model has to be redesigned. Fields of application are multiple. An example of application that will be used as testbed is ambient assisted living, where the process is addressed to model the activities used to help people living an independent life at their home even when they experience physical and mental diseases.
KEYWORDS: Ambient assisted living, process mining, process evolution.
Multiple process interaction discovery
This thesis address the issue of discovering processes interaction inside an organization and between multiple organizations. These interactions are not explicit most of the time. Discovering them is important for better understanding the successes and faults in the organization management. The analysis of interaction will start from the data produced during the processes execution (logs, events) and from the data available in the environment where the organization operates which can affect the results. At the end of the discovery process it will be possible to perform a deeper analysis of the organization results and to answer complex questions. This analysis can be applied to ambient assisted living in order to answer questions like: (i) which are the activities followed by the production of a report? (ii) Which is the average duration of an activity for a specific subject? Is it regular? Does it change with operator performing it? Is it in line with the duration of the same activities on other subjects? (iii) Does the room temperature interfere with a specific activity? How? Which actions can be taken to avoid it?
KEYWORDS: Ambient assisted living, process mining, complex event processing.