Monday, September 29, 2025, 10:00-11:15
Keynote Session 1
Conference Hall

Moderator: Vladimir Voevodin

Conference Opening
Vladimir Voevodin
Moscow State University, Russia

Current challenges in developing high-performance computing technologies, including supercomputer modeling, for the development of high-tech products
Rashit Shagaliev
Deputy Director of RFNC-VNIIEF - Deputy Scientific Director - First Deputy Director of the Institute of Theoretical and Mathematical Physics, Doctor of Physical and Mathematical Sciences, Corresponding Member of the Russian Academy of Sciences
Andrey Grebennikov
Deputy Head of the Mathematics Department of ITMP, RFNC-VNIIEF, Doctor of Physical and Mathematical Sciences

High-Performance Systems for AI Tasks: How Technologies and Trends Are Changing Approaches to Construction and Operation
Alexander Naumov
Director of the Supercomputing and Client Infrastructure Development Department at Cloud.ru
Sergey Kovylov
Chief Architect of Application Solutions at the Supercomputing and Client Infrastructure Development Department at Cloud.ru
Oleg Sergeevich Aladyshev
Expert at the Supercomputing and Client Infrastructure Development Department at Cloud.ru

The presentation will discuss current trends in building high-performance systems for artificial intelligence tasks. It will cover requirements, hardware design, software approaches, and system performance under various workloads, from both the end-user perspective and the perspective of the engineering teams creating and maintaining such systems.

FEP-Based Large-Scale Virtual Screening for Effective Drug Discovery against COVID-19 on the Tianhe supercomputer
Chengkun Wu
Associate Professor, National Key Laboratory of Parallel and Distributed Computing, College of Computer Science and Technology, National University of Defense Technology

Monday, September 29, 2025, 11:45-13:30
Keynote Session 2
Conference Hall

Moderator: Vladimir Voevodin

AI technologies in healthcare: infrastructure, data, reliability and quality assessment, implementation (experience and evolution of the Moscow experiment)
Vladzymyrskyy Anton V.
Research and Practical Clinical Center for Diagnostics and Telemedicine Technologies of the Moscow Health Care Department, Moscow, Russia, Deputy Director for Research, MD, PhD

Architecture of a Distributed Computing System with Hardware Processing of Graph Structures and Associative Memory
Aleksey Popov
Doctor of Engineering, Professor at Bauman Moscow State Technical University, Head of the Teragraph Project

Supercomputing, Computational Physics, and Phase Transitions
Lev Shchur
Professor, Laboratory for Computational Physics, HSE University

We will introduce the audience to two algorithms for modeling systems with complex energy landscapes. Both algorithms have two similar words in their names: “population annealing.” However, these algorithms are fundamentally different from each other. The first is a method based on a temperature annealing procedure using thermalization at a new thermostat temperature (or optimization parameter) value. The second is a sequential traversal of the energy spectrum with the implementation of random configuration selection. What these approaches have in common is that they allow parallel modeling of a large number of copies of the system, i.e., the history of population change. Thus, they are ideal for implementation on clusters and supercomputers with hybrid architecture. We will give a few examples demonstrating their unique properties, realized on supercomputer cHARISMa at HSE University.

Presentation by the RSK Group of Companies (title to be confirmed)
Alexander Moskovsky
CEO of RSC Group of Companies

A New Approach to Solving the Problem of Storing and Ensuring the Availability of Large Volumes of Data
Gleb Trofimov
CEO of Smart Archive

Tuesday, September 30, 2025 г., 17:00-17:30
Keynote Session 3
Conference Hall

Moderator: Vladimir Voevodin

Awarding and Conference Closing
Vladimir Voevodin
Moscow State University, Russia