Andrea Rosà I am a Senior Researcher and Lecturer at the Faculty of Informatics of Università della Svizzera italiana (USI) in Lugano, Switzerland. I'm working in the Dynamic Analysis Group (DAG). The goal of my research is understanding and removing performance inefficiencies in large-scale systems, focusing in particular on managed runtimes. My research interests and experience include analysis and optimization of large-scale parallel and distributed systems, concurrent computing frameworks, compiler optimizations, and benchmarking. I received the PhD in Informatics from USI in 2018, advised by Prof. Walter Binder. Previously, I received the MSc and BSc in Computer Science Engineering from Politecnico di Milano.

Personal Information

• Email:                    andrea [dot] rosa [at] usi [dot] ch
• CV:                        [PDF]
• Publication List:     [PDF]
• Profiles:                 [DBLP] [Google Scholar] [ORCiD].
• Address:                Università della Svizzera italiana, Faculty of Informatics, East Campus, Office D5.10, Via la Santa 1, CH-6962 Lugano-Viganello, Switzerland.
• Phone:                  +41 58 666 4739

Selected Publications

• Heap-Snapshot Matching and Ordering using CAHPs: A Context-Augmented Heap-Path Representation for Exact and Partial Path Matching using Prefix Trees [link]
  Matteo Basso, Aleksandar Prokopec, Andrea Rosà, Walter Binder.
  Proc. ACM Program. Lang., Vol. 9, OOPSLA, 2025.
Improving Native-Image Startup Performance [link]
Matteo Basso, Aleksandar Prokopec, Andrea Rosà, Walter Binder.
CGO 2025.
• Optimization-Aware Compiler-Level Event Profiling [link]
  Matteo Basso, Aleksandar Prokopec, Andrea Rosà, Walter Binder.
  ACM Transactions on Programming Languages and Systems (TOPLAS), 45(2), 2023.
• Analysis and Optimization of Task Granularity on the Java Virtual Machine [link]
  Andrea Rosà, Eduardo Rosales, Walter Binder.
  ACM Transactions on Programming Languages and Systems (TOPLAS), 41(3), 2019.
• Automated Large-scale Multi-language Dynamic Program Analysis in the Wild [link]
  Alex Villazón, Haiyang Sun, Andrea Rosà, Eduardo Rosales, Daniele Bonetta, Isabella Defilippis, Sergio Oporto, Walter Binder.
  ECOOP 2019.
• Renaissance: Benchmarking Suite for Parallel Applications on the JVM [link]
  Aleksandar Prokopec, Andrea Rosà, David Leopoldseder, Gilles Duboscq, Petr Tuma, Martin Studener, Lubomìr Bulej, Yudi Zheng, Alex Villazón, Doug Simon, Thomas Würthinger, Walter Binder.
  PLDI 2019.
• Failure Analysis and Prediction for Big-Data Systems [link]
  Andrea Rosà, Lydia Y. Chen, Walter Binder.
  IEEE Transactions on Services Computing, 10(6), 2017.
  
A complete publication list can be found here.

Selected Software

• Renaissance [link]
  Modern, open, and diversified benchmark suite for large-scale parallelism on the JVM.
• JVBench [link]
  Benchmark suite for the Java Vector API.
• P3 [link]
  Profiler suite for concurrent computing frameworks.
• tgp [link]
  Tool for analyzing and optimizing the granularity of concurrent tasks.
• NAB [link]
  Framework for automated large-scale multi-language dynamic program analysis in the wild.
• DiSL [link]
  Program analysis framework for Java bytecode.
• AkkaProf [link]
  Tool for profiling Akka actors in parallel and distributed applications.
A complete list of software I have led or contributed to is available in my CV.

Last update: September 2025.