Keynote Speakers

Catia Trubiani, Gran Sasso Science Institute (GSSI), Italy 


Predicting is better than Repairing, a tale on Software Performance degradation 


Anticipating the performance degradation of complex software systems is not trivial, even more so when systems are exposed to multiple sources of uncertainty, e.g., heterogeneous workloads leading to fluctuations in resource demands. The goal of this talk is to present some methodologies on how to interpret the performance analysis results with the goal of predicting software performance degradation and avoiding expensive repair actions. Recent results on real-world java applications and microservices will be presented, along with challenges and open research directions. 

Short bio:

Catia Trubiani is Associate Professor at the Gran Sasso Science Institute (GSSI), Italy. Among various national and international projects, she recently received funding by the call MUR-PRIN'22 under the Young Line action, and acting as scientific coordinator of the project. Her main research interests include the quantitative modelling and analysis of interacting heterogeneous distributed systems. More information: .

Enrico Vicario, University of Florence, Italy 


Leveraging non-Markovian models for the integration of model-based and data-driven approaches 


Quantitative evaluation of stochastic models plays a crucial role in the development and operation of a wide range of systems, applications, and processes, with notable applications in mastering the intertwined effects of concurrency and timing.

Validity and ease of generation of these models can now benefit from the growing spread of data technologies and data driven approaches, enabling combination of a-priori knowledge acquired by-design or by-contract together with parameters that are better learnt from observations.

To this end, representation of timings beyond the limit of memoryless Markovian behaviour largely improves the expressivity of models and their ability to retain observed or predefined characteristics. However, this also poses a hard challenge to make the same models amenable to efficient and tool-supported solution for practical implementation of quantitative analytics.

This talk will overview the Oris tool and its underlying Sirio library. We will illustrate how this platform can be used in generation and analysis of models that combine timing constraints enforced by-design with stochastic parameters capturing observed behaviour. We will mention the underpinning theoretical principles and further challenges for development of theory and practice quantitative evaluation of non-Markovian models .

Short bio:

Enrico Vicario is a Full Professor of Informatics Engineering, Head of the Department of Information Engineering of the University of Florence, Director of the Sw Technologies Lab (


He works in the area of Software Engineering, with a twofold focus on applications and methods of quantitative evaluation of stochastic models and on the practice of software development.  He is author of more than 150 papers indexed on Scopus.


He serves as chair of the steering committee of QEST and General Chair of ISSRE'23, and has repeatedly served in the Program Committee of major conferences in the area of quantitative evaluation, including QEST, ISSRE, DSN, EDCC, EPEW, FORMATS, ICPE, ECMS, ValueTools.

He is a faculty member of the PhD program in Smart Computing, jointly delivered by the Universities of Florence, Pisa, and Siena  (


He carries a continuous activity of experimentation and technological transfer in the area of SW architecture and engineering, and is cofounder and scientific director of two spin-offs of the University of Florence: Jaewa (, active since year 2013; and Wedge Engineering (, active since February 2022.


He is scientific leader for the development of the Oris tool and Sirio Java library for production and evaluation of stochastic models (