Research
We work on middleware runtimes and technologies to provision applications in the Cloud (also on the Edge), leveraging Machine Learning for "smart orchestration", minimizing resource usage and energy consumption, while boosting Quality of Service.
Research at the group can be roughly divided into four lines:
1. Methods for AI Workload and Storage Orchestration
- Platforms for Data Analytics in the Cloud
- Virtualized, Containerized Workloads and Data Storage
- AI-driven Edge-Cloud Orchestration
People involved: Josep Ll. Berral, Ramon Nou, Peini Liu, Erin Call, Marc Palacín, Joan Oliveras, David Aguilera.
2. Green Computing in the Cloud/Edge
- Energy-Aware Systems and Data-Centres
- Virtualization and Containerization Technologies
- Heterogeneous Workloads
People involved: Jordi Guitart, Mengxue Wang, Julita Corbalán.
3. Emerging HPC Technologies for Artificial Intelligence
- Scalability of AI in Supercomputing
- Accelerators and Enablers for AI
- Distributed Learning, Federated & Swarm
People involved: Jordi Torres, Alberto Gutiérrez, Pol García Recasens, Ferran Agulló.
4. HPC for Deep Learning, Generative Models and Multimedia
- Unsupervised Image-to-Image Translation with Generative Models
- Artificial Intelligence for Animation
- HPC for Scalable Numerical Simulations
People involved: Ruben Tous, Octavio Castillo.