| Abstrakt: |
The paper deals with the formation of system interfaces of a digital platform to manage the process of providing scientific services, including integration of scientific services to solve a complex research problem, for example, the multiscale modeling of the properties of new materials or conducting interdisciplinary research. Based on the functional structure of the integration of the scientific services of a digital platform (as applied to multiscale modeling) presented in this study, the requirements for the system interface are formulated, and the architecture of the system interface for the integration of scientific services of the digital platform is proposed. The proposed model of a system interface for informational interaction with the services and datasets of a digital platform in management processes for the provision of scientific services is based on modern solutions for managing a virtual infrastructure, based on container processing technologies and microservices, as well as means of orchestration and communication of containers (service mesh) and agile integration technologies. Taking into account the fact that the main function of the digital platform is to ensure the processes of preparation and conduct of research by formalizing the interaction scenarios of researchers, suppliers (sources) of the initial data, and consumers of the results, together with tools for supporting system interfaces to the catalogs of the digital platform, the project offers tools for organizing the interaction of services registered on the platform in order to ensure the execution of scientific research. Synchronization of the processes of providing services, ensuring the transfer of data between services, and obtaining the final result are also ensured through the implementation of control processes of the digital platform, which is based on the proposed system interface. The developed model of system interfaces is new in this work. The proposed interaction interface allows us to effectively consolidate high-performance computing resources and mathematical models based on digital platform technologies. This is especially important for organizing the solution of multiscale modeling problems as a kit of models, each of which operates on the same space-time scale. [ABSTRACT FROM AUTHOR] |
Full text from SpringerLink