STRATEGY STUDY FOR LOAD BALANCING IN DISTRIBUTED DATA PROCESSING SYSTEMS 

Bershadsky Alexander Moiseevich, Doctor of engineering sciences, professor, head of sub-department of CAD systems, Penza State University, bam@pnzgu.ru
Kurilov Leonid Sergeevich, Candidate of engineering sciences, associate professor, sub-department of CAD systems, Penza State University, kleon@gmail.ru
Finogeev Aleksey Germanovich, Doctor of engineering sciences, professor, sub-department of CAD systems,
Penza State University, finogeev@sura.ru

681.3

This paper deals with methods of load balancing in distributed systems. Taxonomy of load balancing strategies is given. After discussing traditional approach to load balancing problem the prognostic strategies are studied. The model of forecast for future load based on fractal method is proposed. It is concluded that most optimal way to solve complicated load balancing problem is to use both traditional and fractal approaches making prognosis.

distributed network systems, load balancing, strategies taxonomy, prog-nostic strategies.