How to use Quantum ESPRESSO on new GPU based HPC systems

How to use Quantum ESPRESSO on new GPU based HPC systems

08May2020

New HPC clusters based on accelerators will soon enter the production phase. These machines are conceptually different from most of those previously  employed  by Quantum ESPRESSO users. The MaX European Centre of Excellence is working to prepare Quantum ESPRESSO for the new and forthcoming architectures. This webinar will provide an update on the status of the GPU version of Quantum ESPRESSO and on the roadmap for its future evolution. It will also provide a basic set of instructions on how to tune and use the code efficiently on these new HPC systems. 

Quantum ESPRESSO is a collection of codes for electronic structure computations widely used in the materials research community. Its HPC users are typically accustomed to machines based on massive MPI parallelism and often need to acquire more familiarity with heterogeneous machines based on GPUs.  The webinar will provide a broad view of the different versions of the Quantum ESPRESSO code, where to get them and where to use them. Following a general introduction, the webinar will focus on how to obtain an optimal performance from Quantum ESPRESSO on these new systems. An introductory talk will provide basic instructions on how to manage the resources on GPU based machines, followed by a tutorial explaining how to compile and tune up our codes.  

   

Who Should attend

Quantum ESPRESSO users, particularly those who plan to start working on new heterogeneous systems. The second part of the webinar is planned to provide more specific information needed by users with allocated compute time on heterogeneous HPC systems (PRACE, ISCRA, etc...).  

 

Key takeaways

  • Status and features of Quantum ESPRESSO running on GPU based systems
  • How to use optimally Quantum ESPRESSO on heterogeneous HPC systems
  • How to compile and tune-up Quantum ESPRESSO for GPUs, which tools and libraries are needed.

 

Agenda

15:00 - 15:15Introduction & Quantum ESPRESSO towards exascale - Pietro Delugas, SISSA
15:15 - 15:30Introduction to GPU-accelerated architectures for HPC: the case of Marconi100 - Fabio Affinito, CINECA
15:30 - 15:45Quantum ESPRESSO on Marconi100: Best Practices, Tips & Tricks - Pietro Bonfá, Uni Parma & CNR Nano
15:45 - 15:55Q&A Session
15:55 - 16:00Closing Remarks

 

This MaX webinar series

This is the first webinar of a series presenting the most recent developments of the MaX flagship codes. The next planned appointments are for AiiDA (May 27), Yambo (June 16), and CP2K (June 24). Stay tuned for the next announcements! 

 

REGISTER HERE

 

Publication date: 08May2020