Research project: "Parallel Algorithms for Large-Scale Numerical Simulations"
Goal
The overall goal of this project is to develop fast parallel matrix solvers for the next-generation parallel computers that enable to reduce significantly the overall simulation time.
Objectives
The objectives of this project are:
- To devise new numerical linear algebra methods for high-performance computers.
- To develop efficient parallel programming techniques for distributed memory computers and hardware accelerators such as GPUs.
- To develop numerical libraries and software for high-performance computers.
Topics
- Numerical linear algebra
- High-performance computing
- Parallelism
- GPU programming
Key technologies
- Iterative matrix solvers and preconditioners
- Distributed memory
- GPU computing
Applications
- Fusion research
- Computational Fluid Dynamics
- Computational Electromagnetics
- Cardiac simulations
Contact
Bruno Carpentieri (
, 
web site)
, web site)Bruno Carpentieri is an associate professor at the Faculty of Computer Science of the Free University of Bozen-Bolzano. Find out more about him checking out his website.
Category
Scientific computing
Type
Research project
