Paris, April 28, 2020
Two of the most potent supercomputers in France, Joliot-Curie, operated at the CEA’s supercomputing centre (TGCC), and Occigen, operated at CPU’s supercomputing centre (CINES), are delivering urgent computing access to huge computer system methods to European research teams associated in the battle towards COVID-19. The goal is to conduct epidemiological scientific tests of COVID-19 virus spread, have an understanding of its molecular composition and behavior and massively screen and exam possible potential molecules, to speed up the look for for an helpful vaccine and assistance the worldwide battle towards the virus. Equally supercomputers are based mostly on Atos’ BullSequana system, worldwide leader in electronic transformation.
Following only a couple months of GENCI’s COVID 19 rapidly keep track of access, GENCI, the French national superior-effectiveness computing agency, announces that there are now a lot more than twenty scientifically assorted COVID19 assignments, managing on its 3 national supercomputers among which are Joliot-Curie and Occigen, with each other with Jean Zay at IDRIS, with the focused assistance of assistance teams in these facilities.
Joliot-Curie at TGCC
Joliot-Curie, with its 22 petaflop/s, is the most potent supercomputer in France focused to educational and industrial open up research. As portion of PRACE (Partnership for State-of-the-art Computing in Europe), European experts are relying on Joliot-Curie to run these days 3 COVID-19 huge scale assignments in the context of the PRACE COVID-19 Rapid Track phone. Just one undertaking simulates the practical proteins of the SARS-CoV-2 virus designed from hundreds of thousands of atoms to have an understanding of the mechanisms of the virus an infection in buy to acquire therapeutics. The other undertaking takes advantage of computational screening, a nicely-recognised technique used in drug discovery, to determine and make improvements to viral protein inhibitors, molecules which are able of blocking the SARS-Cov-2 protein. This undertaking could assistance supporting the advancement of a cure to slow down the COVID-19 outbreak.
The last undertaking brings together the analyze of the influence of antimalarial drugs on a variety of human coronary heart pace styles having into account a assortment of comorbidities that may well be current in the infected population, and also takes advantage of computational fluid dynamics (CFD) to much better have an understanding of the intricate hemodynamics linked with Nord-South Syndrome.
In addition to these PRACE COVID-19 Rapid Track assignments, teams of scientists are carrying out preparatory simulations on precedence national allocations of computing capability on the Joliot-Curie supercomputer. For case in point, scientists from the Grenoble Interdisciplinary Investigation Institute (IRIG) and the CEA’s Joliot Institute are functioning on the look for of inhibitors in COVID-19. The SPIKE protein will allow the virus to penetrate the mobile membrane. Thanks to the simulation of the digital composition of the protein and the linked inhibitor, it is possible to present exact details on the power of the inhibition but also structural details to determine the amino acids worried and their linked polarities. This initial operate made it possible to validate the tactic and therefore to post a 15-million-hour undertaking to PRACE to analyze the microscopic and thermodynamic factors that may well or may well not favour the conversation amongst the most important SARS-CoV-2 protease and promising new inhibitors. The aim is to build an ab initio in silico instrument to estimate accurately the conversation houses of proteins interacting with all styles of ligand households.
Occigen at CINES
Amongst the COVID-19 related assignments on Occigen, scientists are managing simulations to analyze the SARS-Cov-2 helicase enzymes in additional detail, in buy to much better have an understanding of the genetic make-up of the virus. It is also currently being massively used – each day outside of forty,000 cores, about 50 percent of the complete Occigen’s capability formed by 86,000 cores for 3.five petaflop/s – with computational screening procedures, to virtually exam a lot more than one.five billion molecules a thousand of which will be synthesized and tested in labs for their means to inhibit SARS-Cov-2. The scale of this virtual screening is unparalleled and is only made possible thanks to the computing electric power of the supercomputer.
Stephane Requena, CTO of GENCI, explained: “We’re very pleased with the 3 national facilities to be in battle towards COVID-19 and that our potent supercomputing and AI abilities are currently being used by scientists to lead to these initiatives. We hope that these computational findings from the twenty+ assignments at this time managing on our programs, will assistance scientists and experts to have an understanding of a lot more about the virus and in the long run assistance acquire a cure to quit this pandemic.”
Pierre Barnabé, Head of Community Sector & Protection and Head of Significant Information & Cybersecurity, Atos, included: “We are honored to assistance those people scientists and experts in their operate to assistance battle towards COVID-19. By employing the capability of our BullSequana supercomputers, which run 1000’s of periods speedier than typical computers, we are enabling them to save important time and assisting obtain a get rid of to this pandemic.”
Christine Ménaché, Head of the CEA’s Quite Massive Computing Centre (TGCC) and Boris Dintrans, Director of CINES jointly condition: “All the TGCC’s and CINES’s assistance teams have been mobilized in a potent spirit of solidarity to meet up with the expectations of all people and in unique these scientific assignments in connection with Covid19.”
Supercomputers have tens of 1000’s of processors that operate with each other to conduct huge calculations, system and review huge amounts of facts employing AI algorithms. GENCI’s completely combined supercomputing programs characterize these days a lot more than 41 petaflop/s of computing capability and 100 petabytes of enormous facts storage.