Models to support design of highly reliable distributed computer systems with redundant processes of data transmission and handling

In today's economic conditions, engineers are faced with the problem of designing complex distributed computer systems under tight deadlines. Ensuring high reliability and robustness of such systems requires redundancy of their structure as well as the processes of data transmission and handling. The purpose of this study is to create simulation models and means to support computer-aided design of highly reliable distributed computer systems. The simulation environment is AnyLogic 7, used in conjunction with Microsoft SQL Server 2008. The developed models and means are aimed at the optimization of design solutions to the arrangement of the processes of redundant multipath distribution of copies of requests and their redundant handling by clustered servers, interconnected through the given network, under conditions of failures, transmission errors, losses of requests, and limitations on the acceptable residence time of requests in the system. The possibilities of the proposed means for design support, for example choosing the optimal redundancy order of data transmission and processing, are presented. The scope of the efficient use of redundant distribution and handling of requests is determined.