Intellectual computer mathematics system inparsolver
Khimich A.N.,Chistyakova T.V.,Sydoruk V.A.,Yershov P.S.,,,,
DOI: https://doi.org/10.15407/jai2020.04.060
IF: 14.4
2020-12-25
Artificial Intelligence
Abstract:The paper considers the intellectual computer mathematics system InparSolver, which is designed to automatically explore and solve basic classes of computational mathematics problems on multi-core computers with graphics accelerators. The problems of results reliability of solving problems with approximate input data are outlined. The features of the use of existing computer mathematics systems are analyzed, their weaknesses are found. The functionality of InparSolver, some innovative approaches to the implementation of effective solutions to problems in a hybrid architecture are described. Examples of applied usage of InparSolver for processes mathematical modeling in various subject areas are given. Nowadays, new more complex objects and phenomena in many subject areas (nuclear energy, mechanics, chemistry, molecular biology, medicine, etc.) are constantly emerging, which are subject to mathematical research on a computer. This encourages the development of new numerical methods and technologies of mathematical modeling, as well as the creation of more powerful computers for their implementation. With the advent and constant development of supercomputers of various architectures, the problems of their effective use, expansion of tasks range should be solved, ensuring the reliability of computer results and increasing the level of intellectual information support for users ‒ specialists in various fields. Today, the issues of solving these problems are given special attention by many specialists in the fields of information technology and parallel programming. The world's leadingscientists in the field of computer technology see the solution to the problems of efficient usage of modern supercomputers in algorithmic software creation that easily adapts to different computer architectures with different types of memory and coprocessors, supports efficient parallelism on millions of cores etc. In addition, improving the efficiency of high-performance computing on modern supercomputers is provided by their intellectualization, transferring to the computer to perform a significant part of the functions (symbolic languages for computer problem statement, research of mathematical models properties, visualization and analysis of tasks results, etc.). The industry of development and usage of intelligent computer technologies is one of the main directions of science and technology development in modern society
computer science, artificial intelligence