Forecasting the efficiency of multicomponent computer systems

Fìz.-mat. model. ìnf. tehnol. 2021, 32:96-100

Abstract

The advantages and disadvantages of Mooreʼs, Hilderʼs, Amdalʼs, Gustafson-Barsis laws known in the field of information and communication technologies are shown, offering the necessary mathematical apparatus for constructing similar laws for predicting the efficiency of modern multicomponent computing systems. This apparatus includes both the parameters of the components of computer systems, and possible interdependencies between those parameters. In general, forecasting the efficiency of computer systems requires detailed documentation of the work of computer systems of the class given (the series given) on certain types of tasks with subsequent processing of the data obtained. The collection and processing of this data must take place and be recorded in the dynamics with assistance of appropriate smart sensors of the Internet of Things class.

References

1. Gorbachuk, V. M. (2008). Parallelnyye metody resheniya zadach dvukhurovnevogo programmirovaniya i ikh primeneniya. Kompyuternaya matematika, 1, 159–165. (In Russian).
2. Sergienko, I. V., Khimich, O. M. (2019). Matematychne modeliuvannia: vid MELM do ekzaflopsiv. Visnyk Natsionalnoi akademii nauk Ukrainy, 8, 37–50. (In Ukrainian).
3. Amdahl, G. (1967). Validity of the single-processor approach to achieving large-scale computer capabilities. AFIPS Conference Proceedings, 30, 483–485.
   DOI doi.org/10.1145/1465482.1465560
4. Arthur, W. B., Ermoliev, Yu. M., Kaniovski, Yu. M. (1987). Path-dependent processes and the emergence of macro-structure. European Journal of Operations Research, 30, 294–303.
   DOI doi.org/10.1016/0377-2217(87)90074-9
5. Gustafson, J. L. (1988). Reevaluting Amdahlʼs law. Communications of the ACM, 31(5), 532–533.