The Implementation of the Genetic Algorithm Using Cloud-Based Computing on the Internet

Two important tasks are considered in this paper: highlighting extreme subsets and finding cliques of a graph. These tasks have a great practical importance in the design of automation devices and computing equipment, as well as the creation of control systems, computers and networks, in sociology and game theory. Because they are belong to the class of NP-hard problems, the study, creation and modification of methods for their solution still remains relevant. In connection with the expansion of the information space and the increase information for analysis in times, it becomes obvious that in practice it is advisable to solve such NP-difficult problems using new IT technologies - cloud computing. In this regard, the fundamental difference of this work is not only the construction and verification of the genetic algorithm modification, but also the use of the cloud platform as an Internet service for performing calculations. The main goal of a research is to solve the problem of finding extreme subsets in a graph and building cliques using modified genetic algorithm in a reasonable time. The principal difference of the proposed method is that it considers the population degradation degree and increases the diversity of the population using various genetic operators in the evolutionary adaptation block. Several experiments were also conducted, during which a comparison of the developed algorithms with various crossovers was carried out: ordering crossover and greedy. The practical results of the research almost coincided with the theoretical background.