This article examines the concept, architecture, and implementation principles of an integrated computing environment (ICE) for high-performance modeling and machine learning on next-generation supercomputers. The core of this software is the Basic Modeling System (BMS), which includes modern algorithmic and technological support for all key stages of solving large-scale applied problems, as well as functional support for optimization methods for inverse problems implementing the principles of intelligent innovation. Overall, the ICE is a multi-level system with a wide, evolving set of mathematical tools, including automated algorithm parallelization and the creation of flexible application configurations, domain-specific active knowledge bases, problem and computational data archives, scenario designers and analyzers, decision-making tools, and more. The project is focused on a long lifecycle and effective applications, which should ensure flexible expansion of the implemented models and methods, adaptation to the evolution of computer platforms, the reuse of external products, and the coordinated participation of various development teams.
V. P. Ilyin (Thu,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: