LADE – Data Engineering Laboratory
The Data Engineering Laboratory manages and analyses data generated by the genomics and microscopy laboratories (LAGE and LAME), and carries out research and development in the field of artificial intelligence to extract knowledge and value from data. The core element is ORFEO, a datacenter that manages and analyses scientific data in an open-source framework.
ORFEO (Open Research Facility for Epigenomics and Other) is a modern datacenter that provides advanced computing and data-analysis services. The system is capable of delivering over 4 million hours of computing per year. The system is highly flexible and can be reconfigured: alongside the computing servers there is an infrastructure enabling HPC-AI services to be provided via the cloud.
The datacenter also contributes to processing data generated by the Life Sciences Platform and the Innovative Materials Platform.
Technical Features
The container hosting the data centre is connected through redundant 10 Gbit fibre optics to the network of Area Science Park and Lightnet, offering fast connection speeds with the entire regional and national research network. The infrastructure, designed with full redundancy of all components, offers a set of high-memory computing nodes (up to 1.5 TB of RAM per server) and latest-generation processors. Certain servers are equipped with dual V100 GPGPU cards with 32 GB of RAM.
The total storage capacity reaches 3 PB of space in raw mode and is served to users through a CEPH parallel file system for the fast SSD-based component and a standard component. The long-term storage partition is over 1.5 Petabytes in size.
LABORATORY CHIEF: Stefano Cozzini
Latest LADE publications:
- T. Rodani, E. Osmenaj, A. Cazzaniga, M. Panighel, C. Africh, S. Cozzini*, Towards the FAIRification of Scanning Tunneling Microscopy Images, 2023
- L. Valeriani, F. Cuturello, A. Ansuini*, A. Cazzaniga*, The geometry of hidden representations of protein, 2022
- F. Cuturello^, F. Pozzo^, E. N. Villegas Garcia, F. M. Rossi, M. Degan, P. Nanni, I. Cattarossi, E. Zaina, P. Varaschin, A. Braida, M. Berton, L. Zannier, F. Vit, E. Tissino, T. Bittolo, R. Laureana, G. D’Arena, L. Laurenti, A. Tafuri, J. Olivieri, F. Zaja, A. Chiarenza, M. I. Del Principe, R. Bomben, A. Zucchetto, S. Cozzini, A. Ansuini, A. Cazzaniga*, V. Gattei*, An Unsupervised Machine Learning Method Stratifies Chronic Lymphocytic Leukemia Patients in Novel Categories with Different Risk of Early Treatment, 2022
- E. T. Russo^, F. Barone^, A. Bateman, S. Cozzini, M. Punta*, A. Laio*, DPCfam: Unsupervised protein family classification by Density Peak Clustering of large sequence datasets, 2022
- D. Doimo, A. Glielmo, A. Ansuini, A. Laio, Hierarchical nucleation in deep neural networks, 2020
^ = contributed equally
* = corresponding author