‘Meet the Researcher’ is a series in which we spotlight different researchers in academia who use NVIDIA technologies to accelerate their work.
Overview
The article features Marco Aldinucci, a Full Professor at the University of Torino, who is at the forefront of research integrating High-Performance Computing (HPC) and Artificial Intelligence (AI) to combat COVID-19. It highlights his contributions, ongoing projects, and the establishment of a national lab focused on HPC technologies.
What You'll Learn
1
How to modernize HPC applications for AI integration
2
Why collaboration with domain experts enhances research outcomes
3
When to apply new methodologies for portable HPC applications
Prerequisites & Requirements
- Solid grounding in computer science and mathematics
- Experience in parallel computing and AI(optional)
Key Questions Answered
What is Marco Aldinucci's role in HPC and AI research?
Marco Aldinucci is a Full Professor at the University of Torino and Director of the HPC Key Technologies and Tools national lab. His research focuses on the convergence of HPC and AI, particularly in applications related to COVID-19, where he leads efforts to develop tools for AI-assisted diagnosis.
What challenges does Aldinucci face in his research?
Aldinucci faces the challenge of keeping up with rapidly evolving technology and paradigms in HPC and AI. He emphasizes the importance of a strong foundation in computer science and mathematics to navigate these changes effectively.
How does Aldinucci's work impact the research community?
Aldinucci's work impacts the community by establishing the HPC Key Technologies and Tools lab, which affiliates researchers from 35 Italian universities. This initiative aims to enhance collaboration and competitiveness in international research projects, gathering significant funding for HPC advancements.
What future projects is Aldinucci working on?
Aldinucci is developing two software projects: StreamFlow, a Workflow Management System for portable HPC applications, and CAPIO, a system for efficient data transfer between parallel applications. These tools aim to enhance the performance and portability of HPC applications.
Key Statistics & Figures
Research funding attracted
over 6M€
Aldinucci has attracted significant research funds to the University of Torino through participation in over 30 EU and national research projects.
Number of researchers affiliated
hundreds
The HPC Key Technologies and Tools lab co-affiliates researchers from 35 Italian universities to enhance collaboration.
Total cost of EU research projects in first year
95M€
In its first year, the HPC-KTT lab gathered EU research projects in competitive calls totaling 95M€.
Number of cores in HPC4AI data center
almost 5000 cores
The HPC4AI center at the University of Torino is equipped with nearly 5000 cores and 100 GPUs.
Technologies & Tools
Some links below are affiliate links. We may earn a commission if you make a purchase.
Hardware
Nvidia
Used to accelerate machine learning tasks in Aldinucci's research.
Cloud Platform
Openstack
HPC4AI runs an OpenStack cloud to facilitate research in data-centric applications.
Container Orchestration
Kubernetes
StreamFlow is designed to be portable across cloud systems using Kubernetes.
Key Actionable Insights
1Engage in multi-disciplinary collaboration to enhance research outcomes.Working with domain experts allows researchers to apply HPC and AI effectively across various scientific fields, leading to innovative solutions and improved research quality.
2Adopt modern AI toolkits to modernize HPC applications.Integrating AI concepts into HPC programming can significantly improve application performance and portability, making it easier to develop scalable solutions.
3Stay updated with technological advancements to remain competitive in research.The rapid evolution of technology necessitates continuous learning and adaptation to new paradigms, which is crucial for success in HPC and AI research.
Common Pitfalls
1
Failing to keep up with rapid technological changes can hinder research progress.
Researchers must continuously learn and adapt to new tools and paradigms to remain effective in their work. This requires a proactive approach to education and skill development.
Related Concepts
High-performance Computing (hpc)
Artificial Intelligence (ai)
Parallel Computing
Cloud Computing