Results for distributed computing network modeling
The article was published in issue №2
The paper presents an overview of platforms for simulation of distributed computing networks. The au-thors have chosen a modeling platform that has the required functionality to study the problem of fault tolerance of a distributed job management system, as well as to study the problem of scheduling user jobs for geo-graphically distributed resources. There is an investigation on the GridSim modeling platform for building a distributed network model. Two models were implemented: one is for investigating fault tolerance of a distributed network job management system problem, and the other one is for investigating auction meth-ods for scheduling user jobs for geographically distributed resources. The authors have formulated the ad-vantages and disadvantages of distributed network modeling. The paper presents an alternative approach requiring less computational resources.