PSI (Pervasive Self-Optimizing Computing Infrastructures) aims to provide a fabric of software components able to dynamically optimize the behaviour of the cloud and IoT infrastructures. PSI targets an improved usage of resources at the systems’ edge via continuous and distributed system-wide optimizations.
This project covers both theoretical and practical aspects, and it combines different research areas including self-adaptive software, control theory, optimization, distributed and real-time systems. The main goal of the project is to develop new methodologies for the efficient usage of computational resources while providing guarantees on different key performance indicators, like, for example, response time and throughput of the system.
Optimal Reference Tracking for Sampled-Data Control Systems (Dec 2022) Enrico Bini , Alessandro Papadopoulos, Jacob Higgins , Nicola Bezzo 61st IEEE Conference on Decision and Control (CDC 2022)
Kubernetes Orchestration of High Availability Distributed Control Systems (Aug 2022) Bjarne Johansson, Mats Rågberger , Thomas Nolte, Alessandro Papadopoulos 23rd IEEE International Conference on Industrial Technology (ICIT 2022)
RT-SCALER: Adaptive Resource Allocation Framework for Real-Time Containers (Jul 2022) Václav Struhár, Silviu Craciunas , Mohammad Ashjaei, Moris Behnam, Alessandro Papadopoulos Real-time And intelliGent Edge computing workshop (RAGE2022)
Feedback-based resource management for multi-threaded applications (Jul 2022) Alessandro Papadopoulos, Kunal Agrawal , Enrico Bini , Sanjoy Baruah Real-Time Systems (RTSJ 2022)
Optimization-based attack against control systems with CUSUM-based anomaly detection (Jun 2022) Gabriele Gualandi, Martina Maggio , Alessandro Papadopoulos The 30th Mediterranean Conference on Control and Automation (MED 2022)
STRETCH: Virtual Shared-Nothing Parallelism for Scalable and Elastic Stream Processing (Jun 2022) Vincenzo Gulisano , Hannaneh Najdataei , Yiannis Nikolakopoulos , Alessandro Papadopoulos, Marina Papatriantafilou , Philippas Tsigas IEEE Transactions on Parallel and Distributed Systems (TPDS)