Ðóñ Eng Cn Translate this page:
Please select your language to translate the article


You can just close the window to don't translate
Library
Your profile

Back to contents

Software systems and computational methods
Reference:

Grishentsev A.Yu., Korobeinikov A.G. Problem definition for optimization of distributed computing systems

Abstract: the article describes a model and problem definition for optimization of distributed computing systems. The results of the study are in good accordance with Amdahl’s law and together with the game theory and optimizations allow finding the most successful solutions in terms of efficient use of computing resources when designing or upgrading the distributed computing systems. The article discusses the threaded model of distributed computing systems of continuous time. The disadvantage of this model is the possibility of simulating only threaded distributed computing system while the case of transferring of blocks of data requires the discrete time system model. Modern distributed computing systems may contain multiple separate computing units linked through communications network and located in diff erent parts of the Earth and near-Earth space. The authors review block model of distributed computing discrete time system. Such model allows examining both threaded and blocks data processing and considering time of delay needed for data synthesis and transfer. The solution of optimization task can be found by sequential search with appliance of game theory and optimizations for computational tasks, resource nodes.


Keywords:

threading model, distributed computing system, optimization, Amdahl’s law, discrete time system, direct graph, graph node, block model of distributed computing systems, computation channel time, game theory


This article can be downloaded freely in PDF format for reading. Download article

This article written in Russian. You can find original text of the article here .
References
1.
2.
3.
4.