arXiv subjects
Publications and source records attributed to Igor Ivanisenko.
The work presents results of a numerical study of fractal characteristics of multifractal stream at addition of stream, which does not have multifractal properties. They showed that the generalized Hurst exponent of total stream tends to one of original multifractal stream with increase in signal/noise ratio.
The load-balancing system, built on the basis of a subsystem load balancer and subsystem control and monitoring that closely interact with each other was propose in work. This system is presented as a queuing system with priority service discipline. In the described queuing system parallel processing flow applications occur in the multiple serving devices and successive junction of them into unified stream is done. The method of multifractal load balancing is submited on the basis of the developed system of load balancing.
The method of calculating a distributed system imbalance based on the calculation of node system load was proposed in the work. Calculation of node system load is carried out by calculating the average coefficient of utilization of CPU, memory, bandwidth of each system node, taking into account multifractal properties of input data flows. The simulation of the proposed method for different multifractal parameters was conducted. The simulation showed that the characteristics of multifractal traffic have a considerable effect on the system imbalance. Using the proposed method allows to distribute requests across the servers so that the deviation of the load servers from the average value was minimal, which allows to get higher system performance metrics and faster processing flows.
The dynamic load balancing algorithm based on the monitoring server load, self-similar characteristics of passing traffic have to provide a statistically uniform load distribution on servers, high performance, fault tolerance and capacity, low response time, the amount of overhead and losses was propose in work. Integrated measurement for the total imbalance level of the system were entered.
The classification of the most used load balancing algorithms in distributed systems (including cloud technology, cluster systems, grid systems) is described. Comparative analysis of types of the load balancing algorithms is conducted in accordance with the classification, the advantages and drawbags of each type of the algorithms are shown. Performance indicators characterizing each algorithm are indicated.
The work presents results of numerical study of self-similar properties of additive data traffic. It is shown that the value of Hurst exponent of total stream is determined by the maximum value of Hurst exponent of summed streams and the ratio of variation coefficient of stream with maximum Hurst exponent and other ones.