The article describes a method for solving the main problem of backward analysis using specialized digital data conversion algorithms, with open architecture. The proposed method significantly reduces the requirements for computing power and allows you to use standard configuration computers to solve the main problem of backward analysis. A special feature of the method is the ability to separate and make an individual research of the methodological and processor components of the computational errors of transformations, as well as evaluation of the metrological level of the algorithms that implement them. Examples of solving reverse analysis problems are given.

The proposed method and its fragments can be used for conducting computational experiments to investigate the fine properties of computational algorithms used in engineering and scientific computational experiments.

Categories and Subject Descriptors: (Primary) C.0 [Computer Systems Organization]: General -- instruction set design; D.3.4 [Programming Languages]: Processors -- compilers, optimization; G.1.0 [Numerical Analysis]: General -- computer arithmetic, error analysis, numerical algorithms (Secondary): forward and backward analysis

D.2.1 [Software Engineering]: Requirements/Specifications -- languages; D.3.4 Programming Languages]: Formal Definitions and Theory -- semantics; D.4.1 Operating Systems]: Process Management -- synchronization.

General Terms: Algorithms

Additional Key Words and Phrases: Normalized number, floating-point standard, relative error, rounding error, rounding, ulp.

References

[1] J.H. Wilkinson The algebraic eigenvalue problem, Nayka, 1970. -565 p.

[2] Voevodin V. V. Parallel calculations. SPB. BHV. Petersburg, 2002 . 608 с.

[3] Slyusarenko A.S. To the Problem of Rounding Errors Evaluation /International Scientific Review of Problems and Prospects of Modern Science and Education /Collection of Scientific Articles. XlV International Correspondence Scientific and Practical Conference (Boston, USA, May 24-25, 2018). Boston. 2018 ), pp.12–26. [Электронный ресурс].Режим доступа: https://.pdf/(дата обращения: 08.01.2020).

[4] Slyusarenko A.S., Dudarev A.S Evaluation to Investigation of the Wilkinson’s Paradox To the Problem of Rounding Errors ///Collection of scientific articles. XLV International correspondence scientific and practical conference “International Scientific Review of Problems and Prospects of modern science and education”,  (Boston, USA, July 24-25, 2018). Boston. 2018, [Электронный ресурс]. Access mode: https://.pdf/(дата обращения: 08.01.2020).

[5] A.S. Slusarenko, G.N. Dyakova, Metrological aspects of computer based measurement systems in radionavigating complexes, Proceedings of the International Conference Days on Diffraction 2009, St.Petersburg, Russia, (2009) рр.61-65.