Instructor: Alessio Roccon

Numero ore: 8

SSD: ING-IND/06

Learning objectives:
This course provides an introduction to the broad field of Parallel Computing (PC). Designed for doctoral students at the beginning of their journey in this area, it equips participants with the foundational knowledge required to effectively integrate parallel computing into their PhD research. The course begins with the fundamental concepts and terminology of parallel systems, followed by an overview of parallel architectures and memory models, including shared-memory and distributed-memory systems. Participants will then explore key programming paradigms—such as MPI, OpenMP, OpenACC, and CUDA—targeting both multi-core CPUs and GPU architectures. In addition, the course covers commonly used performance metrics, scalability models, and typical bottlenecks, including communication overhead and load imbalance. Bridging theory and practice, the course includes hands-on sessions focused on parallelizing simple problems and applying parallel computing techniques to real research problems.

Teaching methods: lectures
Assessment methods: final written report
Additional information: the course can be delivered in English if required