Versions Compared

Old Version 56

changes.mady.by.user Neva Besker

Saved on

compared with

New Version Current

changes.mady.by.user Alessandro Marani

Saved on

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.




The

new Galileo100 Infrastructure

Leonardo Data Centric General Purpose (DCGP) module is upcoming!

Image Added


We are pleased to announce that the new infrastructure Galileo100 will start its pre-LEONARDO Data Centric and General Purpose  will enter the production phase on Monday,  August 9th. The full production is expected in September 2021.

The new Infrastructure is co-funded by the European ICEI (Interactive Computing e-Infrastructure) project, it is a system engineered by DELL and consists of:

  • 528 computing nodes 2 CPU Intel CascadeLake 8260,  with 24 cores each, 2.4 GHz, 384 GB RAM and 36 additional nodes with 2x NVIDIA GPU V100 with 100Gbs Infiniband interconnection
  • 77 computing server OpenStack for cloud computing, 2x CPU 8260 Intel CascadeLake, 24 cores, 2.4 GHz, 768 GB RAM, with 100Gbs Ethernet interconnection.
  • 20 PB of active storage accessible from both cloud and HPC nodes.
  • 5 PB of fast storage for HPC system.
  • 1 PB Ceph storage for Cloud (full NVMe/SSD)

More information are available in the G100 User Guide.

Quantum Computing @ CINECA

From March 1st 2021, CINECA made available, free of charge, computing hours selected from the most advanced and mature quantum computing technologies currently available on the market to all Italian universities and research centers.
In particular, it will be possible to access Quantum annealing resources, the use of which is possible thanks to the recent collaboration agreements between CINECA and the leading company in the quantum annealing sector, D-WAVE, which currently owns one of the most powerful quantum annealers in the world. Through the agreement, CINECA will make calculation hours available to the Italian scientific community both on the new D-WAVE computer, Advantage, which has a QPU of over 5000 qubits, and on the hybrid D-WAVE machines, modern supercomputers that combine the power of quantum annealers with HPC resources, increasing the computing power of the machines to allow users to solve combinatorial optimization problems using a maximum of one million binary variables (equivalent, therefore, to a quantum annealer with one million qubits).
The agreement provides for a total of 55 hours to be distributed monthly between now and the end of the year (approximately 5 hours per month). Each hour is interchangeable with 20 hours of calculation on hybrid systems.
Quantum annealing resources will be awarded upon submission of an ISCRA-C project. Also, it will be possible to access HPC resources for emulation of quantum computing environments.

February 12th.
LEONARDO, which is in the production phase since Summer 2023, is ranked 6th among the most powerful supercomputers in the Top500 List. 

The European supercomputer Leonardo, managed by Cineca, is based on Atos BullSequana XH2000 technology, and features nearly 14,000 NVIDIA Ampere architecture-based GPUs and NVIDIA Mellanox HDR InfiniBand. It delivers 10 exaflops of FP16 AI performance.

In 2024, Leonardo will receive an upgrade  targeting an increase of 100 PFlops of computing capacity, that will introduce a new partition, LISA,  further expanding its capacity and expertise and serving as an invaluable resource for the European scientific community.

More information on LEONARDO's website!More information.