Wednesday 2 December 2020 14:00
High Performance Computing, a key enabler for digital transformation
Demand for accurate, reliable numerical simulations of complex phenomena is increasing exponentially across a variety of scientific and engineering disciplines ranging from mechanics to underground processes, from computational fluid dynamics to electromagnetism and many others. These applications often require high resolution graphics and run computational domains with several million or even billions of unknowns that impel the use of high-performance computing (HPC).
Moreover, the need for fast simulations is even more important today as emerging technologies, such as the Digital Twin, require continuous elaboration and calibration of the models in nearly real time based on the data being collected.
This session at the International CAE Conference will explore the most advanced algorithms and applications available to better leverage the huge computational capabilities of modern HPC systems.
14:00 Keynote Speaker
Leonardo s.p.a. | Carlo Cavazzoni
High Performance Computing, a key enabler for digital transformation
In a recent speech, IBM CEO Arvind Krishna, said that: “digital transformation has been accelerated during the COVID-19 pandemic and ultimately every company will become an AI company." This is not strictly true, but what is true is that every company will have to adopt AI technologies. AI considered in a broad meaning, and we can be more precise saying that every industry will have to applying digital technologies determining a paradigm shift, the value of goods/services move from the exploitation of physical systems to the exploitation of knowledge. AI, computer simulations and other digital technologies are tools to help mining out more knowledge, faster. The more the better. In this scenario HPC is a tool, a tool to process BigData, enable AI and perform simulations, it can accelerate the creation of value thanks to the capability to generate new knowledge and perform more accurate predictions. Whereas computational capacity is a fundamental resource for competitiveness, row computational capacity alone is useless, a crunching device transforming sequences of 1 and 0, into other sequences of 1 and 0. The software is the key to unlock the value. This is why, beside the supercomputer, we need to create the capability to implement applications or improve the already existing one. In the talk I will present how Leonardo with the key contribution of the HPC Lab, intends to implement leadership software tools and computational infrastructure able to add value to the company, and ultimately transform it, to be more digital than physical.
Carlo Cavazzoni (Male) - Born in Formigine (Mo) the 20/05/1970, is presently head of Computational R&D in Leonardo, and director of the Leonardo HPC Lab. Before joining Leonardo in May 2020, he spent more than 20 years in Cineca (Italian supercomputing centre), where he become head of HPC R&D, with responsibility for the evolution and exploitation of the national and European HPC infrastructure. He is a member of the EuroHPC Research and Innovation Advisory Board. He was also Co-PI and Work Package leader in the MaX Centre of Excellence for Exascale computing, and it was deeply involved in many EC projects. From high level education standpoint, he has been graduated in physics at the University of Modena in the 1994, and subsequently he has attained the degree of PhD at the International School for Advanced Studies (ISAS-SISSA) of Trieste in 1998 with a thesis about: “Large Scale First-Principles Simulations of Water and Ammonia at High Pressure and Temperature”. During his PhD, he has studied various problems concerning the implementation and the efficiency of parallel numerical algorithms being used in computer simulations. He collaborates with different user communities to enable applications on massively parallel HPC systems and innovative architecture solutions. In particular he is responsible for the parallel design of the Quantum ESPRESSO suite of codes. He his author and co-author of 90+ peer review articles, including Science, Physical Review Letters, Nature Materials, and many others.
(semi complete list of publications is available on www.researchgate.net/profile/Carlo_Cavazzoni/publications).
Finally, he was a pioneer in HPC architecture designs with two successful HPC prototypes cofounded by PRACE (prace-ri.eu): Eurora and DAVIDE, build in partnership with Eurotech (Eurora) and E4 (DAVIDE).
In particular Eurora was ranked first in the Green500 lists in June 2013.
14:45 Technical session
CINECA | Ivan Spisso
exaFAOM: Overview and preliminary activities
The presents talk aim to give an overview of the recently approved EuroHPC project exaFOAM: Exploitation of Exascale Systems for Open-Source Computational Fluid Dynamics by Mainstream Industry. The Consortium, led by ESI-OpenCFD consists of a well-balanced group of experts to work on the co-design of OpenFOAM targeting (pre)-exascale HPC architectures. Specifically, the HPC activities already carried out by some of the members of the exaFOAM’s Consortium will be presented. Firstly, the HPC benchmark project which aims to create a framework for enabling the consistent and coherent comparison of performances using OpenFOAM technology across different HW platforms and configurations in view of the transition to exascale. Secondly, the activities related to the GPUs enabling of OpenFOAM will be outlined: i) GPU-accelerated OpenFOAM simulations using PETSc4FOAM ii) AmgX GPU Solver Developments for OpenFOAM iii) a CPU-GPU paradigm to accelerate turbulent combustion and reactive-flow simulation.
Acad. Title: PhD.
First Name: Ivan
Last Name: Spisso
Company/Organization: CINECA, SCAI (Super Computing Application Innovation) Department
Current Job Title: HPC specialist for CFD applications
Ivan Spisso was awarded a PhD in Computational AeroAcoustics, University of Leicester, UK., in 2013.
Since 2010, he works as HPC specialist for academic and industrial CFD applications of SCAI (Super Computing Application Innovation) Department at CINECA, Italian HPC Centre located in Bologna, Italy.
He has been the Organizer of the Workshops: “HPC enabling of OpenFOAM for CFD, applications" held in 2012, 2015 2016 @ Casalecchio di Reno, Bologna, Italy.
He has past and on-going experience of running project PRACE Projects based on performance of OpenFOAM on HPC clusters. He has experience with past and on-going EU CFD projects.
He is one of the promoter and author of the recently approves EU Project, exaFOAM: Exploitation of Exascale Systems for Open-Source Computational Fluid Dynamics by Mainstream.
He is actually the chairman of the OpenOFAM HPC Technical Committee (Term: April 2019 - April 2020), www.openfoam.com/governance/technical-committees.php#tm-hpc
High Performance Computing Center, Stuttgart (HLRS) | Amgad Dessoky
EXCELLERAT: paving the way for the evolution towards Exascale
Engineering applications will be among the first exploiting Exascale, not only in academia but also in industry. The industrial engineering field is the area with the highest Exascale potential. For this reason, the EXCELLERAT activity brings together European experts to establish a Centre of Excellence (CoE) in Engineering Applications on HPC with a broad service portfolio, paving the way for the evolution towards Exascale. The aim is to solve highly complex and costly engineering problems, and create enhanced technological solutions –even at the development stage. This is in line with the European HPC Strategy, as implemented through the EuroHPC Joint Undertaking.
To fulfil its mission, EXCELLERAT will focus its developments on six carefully chosen reference applications (Nek5000, Alya, AVBP, TPLS, FeniCS, Coda), which were analysed on their potential to achieve Exascale performance in HPC for engineering. Thus, they are promising candidates to act as showcases for the evolution of applications towards execution on high scale Demonstrators, Pre-Exascale Systems and Exascale Machines.
I am the Technical Manager of the EXCELLERAT project and have been working for the High Performance Computing Center Stuttgart (HLRS) since February 2020. Before that, I was working for about four years as a Ph.D. Researcher in the Institute of Aerodynamics and Gas Dynamics (IAG) at the University of Stuttgart. Between 2012 and 2015, I worked as Ansys Product Manager in Ansys channel partner in Egypt, at the same time, I was Assistant Lecturer and Research Engineer in the Mechanical Power Engineering Department, Helwan University, Egypt. I graduated from the Mechanical Power Department and graduated with a Master's degree from the same department. I submitted my Ph.D. dissertation in Aerodynamics and Aeroacoustics of the wind turbine in [10, 2020].
Fraunhofer SCAI | Silvia Ehrmann
Application-Aware Multigrid Linear Solver for High Computational and Numerical Performance in Industrial Simulations
Linear solvers form the inner core of computational simulations in various fields of engineering. A steady increase in model resolution (i.e., problem size) puts a high demand for highly efficient solver methods to avoid bottlenecks. Yet, the need for considering more and more aspects from complex physical processes in practical simulations poses challenges regarding robustness and reliability of the linear solver methods.
Algebraic multigrid is the method of choice for many diffusion-driven problems. However, a black-box application is hardly ever possible in real-life situations with complex simulations. Instead, a grey-box approach is followed, where the principal linear solver strategy is adjusted to the concrete type of application, possibly also involving some additional physical information. This approach also renders optimizations for massively parallel execution possible. Once settled, the approach can be effectively used without further need for adjustments.
Silvia Ehrmann has got a Master of Science in Mathematics and Physics from the Heinrich-Heine University Düsseldorf. Since 2017, she works as a research scientist in the SAMG-Team at Fraunhofer SCAI, which develops a highly efficient solver for large linear equations. Her work focuses on the usage of the SAMG-solver in generalized eigenproblems for machine learning applications. Furthermore, she contributed to different consultancy projects with industrial partners, for instance, in the field of performance optimization for linear solver software.
GE AVIO SRL | Donato Magarielli
CFD-based aeronautical turbomachinery case study in the EU-funded LEXIS project
The EU-funded LEXIS project (Large-scale EXecution for Industry & Society, Grant Agreement ID: 825532) aims to build an advanced, geographically distributed, HPC infrastructure for Big Data analytics within three targeted pilot test beds: Aeronautics, Weather and Climate, Earthquake and Tsunami. Within the first one led by Avio Aero, the industrial applicability of last generation HPC/Cloud/BD platforms supporting the execution of sophisticated CAE tools to examine the complex flows behavior in aeronautical engine components is under investigation.
From both a digital technology and business perspective, Avio Aero intends to obtain a marked step change: less time-consuming CAE analyses that exploit newly designed, improved or tightly coupled HW/SW components opening the doors to the “real time” design approach.
Aimed at drastically reducing the execution time of fluid dynamics analyses on low pressure turbines, the turbomachinery case study presented here deals with the improvement of the CFD solver TRAF. The development of a newly designed GPU-enabled version of this code is currently underway with promising results in terms of speed up compared with the consolidated one running on solely CPU-based HPC systems.
Electronic engineer with ~13 years’ experience, covering successful technical roles in different professional gyms: Software Designer in Public & Industrial Research, Signaling Engineer in Railways and currently Digital Technology Product Manager in Aviation.
Skilled to leverage the benefits of Digital Technology to address Engineering and Test Center needs & strategies while managing costs & risks, I’ve been working for 6 years at Avio Aero, a General Electric Aviation business.
Comfortable with facing new business challenges in complex environments implementing cutting edge digital solutions, my professional interests are in the areas of Computer-Aided Engineering, Computer-Aided Testing, Product Lifecycle Management, High Performance Computing, Big data management, and Artificial Intelligence.
CINECA | Claudio Arlandini
The Italian way to Exascale
Computing leadership is determining economic leadership. By pushing the leading edge of computation, new technologies and capabilities ultimately become available for industry, in many sectors like aeronautics, pharma, energy, and automotive.
I will present the status of the Leonardo project, and of the Italian Competence Centre (CC) for HPC. Funded by the European Commission (EuroHPC JU) and the Ministry of the University and Research, conceived and managed by the Cineca Interuniversity Consortium in collaboration with INFN and SISSA, Leonardo will be one of the most powerful supercomputers in the world, able to project Italy towards the exascale class of high performance computing for research and innovation.
The Italian CC for HPC is one of the 27 National Centres cofunded by European Commission and the National Governments through the EuroCC project, and coordinated by the CASTIEL support action, with the aim to enlarge adoption of HPC at all levels especially in industry.
Claudio Arlandini got a PhD in physics at the University of Heidelberg.
He has 20 years of experience in the IT industry mostly in High Performance Computing, 10+ years spent at managing strategic projects and business development activities in Public Private Partnerships.
At CINECA he is supporting business development and technological transfer activities towards industries.
As a Project Manager, he got a large experience in EU & national funded projects aimed at fostering industry innovation and digitisation. Business Development, Business Model Innovation, HPC Cloud, Big Data & IoT technologies.
M3E S.r.l. | Carlo Janna
Chronos: a library for sparse linear algebra problems on High Performance Computers
Numerical simulation is a very common tool in a large number of applications, ranging from mechanics, to underground processes, computational fluid-dynamics, electromagnetism and many others. These applications require high resolution with domains consisting of several millions or even billions of unknowns. In this context, the use of supercomputers is unavoidable, and the development of efficient and scalable linear solvers is a central research topic.
Algebraic multigrid (AMG) methods are often the best choice as preconditioners for the solution of large size 3D problems arising from discretized PDEs. The main advantage of AMG is its theoretical optimality, as, in ideal cases, the iterations to converge do not depend on problem size. In tough problems, standard AMG method is far to be optimal with its performance strictly related to good selection of set-up parameters. In this work, we propose Chronos as a massively parallel implementation of a novel AMG framework able to adapt itself to the problem at hand.
Chronos performance is assessed in the solution of several engineering problems including mechanics, CFD and underground applications.
Carlo Janna graduated in Civil Engineering at the University of Padova in October 2nd, 2003 with 109 points over 110 and in the same University he got his PhD defending a thesis entitled "Numerical modeling of the mechanical behavior of regional faults in the geological sequestration of anthropogenic CO2 sequestration". In December 2011 he became assistant professor at the Department ICEA and since June 2020 he is associate professor in the same Department. The main scientific interests concern on one hand the mathematical and numerical modeling of the mechanics of porous media in both saturated and unsaturated conditions with specific applications in subsurface hydrology and petroleum industry, on the other the numerical linear algebra. His main activity is the development and implementation of numerical models based on the Finite Element method for the simulation of subsurface coupled and uncoupled geomechanical and fluid dynamical processes in the exploitation of deep aquifer or reservoir resources. As to the linear algebra, Carlo Janna studies and develops numerical techniques for the solution of large sparse linear systems and eigenproblems and more specifically iterative methods and preconditioners. For sequential computers, he studied and developed several ad hoc preconditioners for the solution to specific problems arising in subsurface simulations. From 2010 to 2012, Carlo Janna joined the HPC research projects PARPSEA (PARallel Preconditioners for large Size Engineering Applications), SCALPREC (SCALable PREConditioners), OPTIDAS (OPTImization and Data ASSimilation) e SPREAD (Scalable PREconditioners for Advanced Discretizations) studying and developing new preconditioners for massively parallel computers. Carlo Janna is author or co-author of more than 100 scientific papers in international refereed journals, books and conference proceedings.
This event is organized by
with the support of