Antonella DI STEFANO

Antonella Di Stefano is Full Professor of Computer Engineering at the Engineering Faculty of the University of Catania (Dept. of Computerand Telecommunication Engineering). She teaches "Object Oriented Programming"  and "Distributed Systems and Big Data".

Research Activities and Scientific Contributions

For several years, research activity has focused on large-scale distributed systems, contributing models, architectures, algorithms, and protocols for the analysis, evaluation, and design of distributed software infrastructures. Research contributions have addressed Grid, Peer-to-Peer, Cloud, Fog, and Edge computing environments, with particular attention to routing, resource discovery, allocation and scheduling, fault tolerance, and Quality of Service (QoS) management.

Current Research Directions

The research activity is centered on QoS-driven adaptive orchestration and resource management in Edge–Cloud continuum systems, where heterogeneity, latency variability, intermittent connectivity, and resource uncertainty require dynamic and autonomous decision-making mechanisms for service placement, execution, migration, and workload coordination.

Current research directions include:

• QoS-driven orchestration of distributed services
• Task offloading strategies in Edge–Cloud environments
• Predictive and proactive service migration
• Adaptive scheduling and resource allocation under dynamic constraints
• Performance isolation in multi-tenant distributed infrastructures
• Service continuity and SLA-aware orchestration in highly dynamic environments
• Adaptive and autonomic mechanisms for distributed orchestration in large-scale systems

Edge–Cloud Orchestration and Adaptive Migration

Research contributions address the design of distributed and adaptive orchestration mechanisms for large-scale infrastructures operating under highly variable conditions.

Main focus areas include:

• distributed orchestration models for Edge–Cloud continuum systems
• intelligent task offloading across heterogeneous computing nodes
• predictive migration strategies for service continuity and performance optimization
• fault-aware and QoS-aware workload coordination
• adaptive service placement under latency, cost, availability, and reliability constraints
• decentralized strategies for adaptive resource management in dynamic infrastructures

These approaches aim at ensuring service continuity, scalability, and SLA compliance in environments characterized by dynamic workload fluctuations, resource variability, and infrastructure degradation.

AI-based Adaptive Decision for Distributed Systems

Research activities also investigate AI-driven adaptive and autonomic decision mechanisms for distributed infrastructures, supporting:

• dynamic service placement and migration decisions
• QoS- and SLA-aware orchestration policies
• adaptive workload distribution across Edge and Cloud resources
• predictive adaptation under uncertain and partially observable conditions
• distributed adaptive orchestration strategies for large-scale infrastructures

These systems combine adaptive heuristics, autonomic approaches, and data-driven techniques to support runtime adaptation in distributed computing environments.

Research Background and Scientific Foundations

Significant scientific results have been produced in the development of models, architectures, and algorithms for mobile-agent systems and their application to distributed environments. Particular attention has been devoted to scalable mechanisms for naming, discovery, and coordination of mobile agents in large-scale systems, supporting resilient operation under mobility, dynamic migration, and failures.

Research activities on Grid systems have resulted in several national and international projects and scientific contributions. In this context, resource scheduling problems have been addressed in both best-effort and QoS-guaranteed scenarios. QoS-aware job allocation strategies, advance reservation patterns, and scheduling mechanisms for workstation clusters have been proposed and integrated into PBS and LSF schedulers, as well as OGSA-compliant middleware platforms such as Globus and gLite.

In Peer-to-Peer environments, research focused on scalable and fault-tolerant QoS-aware discovery protocols and adaptive resource allocation strategies capable of reacting to abrupt resource fluctuations caused by workload variations and failures. These solutions have proven particularly suitable for Cloud environments characterized by stringent scalability and elasticity requirements.

In Cloud computing, research activities addressed workflow management and autonomic resource optimization. A modular Workflow Management System (IWoM – IaaS Workflow Manager) was designed to support QoS-aware workflow execution over heterogeneous IaaS infrastructures. Additional contributions concerned performance isolation in virtualized systems and autonomic user-side models capable of dynamically adapting the mapping between application components, virtual machines, and physical resources, with the aim of optimizing the trade-off between resource utilization and operational cost. Part of this work has been applied to Smart City scenarios.

Among the proposed solutions for distributed routing and scheduling, particular relevance has been given to the bio-inspired “Alienated Ant Algorithm” (AAA), a fully distributed and adaptive QoS-aware algorithm designed to balance routing flows and resource utilization. Unlike traditional Ant Colony Optimization approaches, AAA explicitly addresses scalability, fault tolerance, and adaptive coordination of resources in highly dynamic distributed environments.

The AAA approach provided a fully distributed and adaptive framework for QoS-aware routing and resource coordination in dynamic environments, based on local decision-making and self-organization mechanisms.

The AAA algorithm has been successfully applied to:

• resource allocation in cloud and data center environments
• distributed monitoring systems
• wireless sensor networks and Software Defined Networks (SDN)
• urban logistics and emergency planning scenarios
• intelligent transportation systems and distributed coordination

More recent activities focus on orchestration strategies for containerized services and intelligent transportation systems.