Management Model in Remote Application Servicing Systems with Conditional Priority
Annotatsiya
The widespread use of information and communication technologies in service systems has led to the emergence of remote service. A distinctive feature of this service is the multi-stage nature of request processing. The increase in the number of types of remote request processing leads to a sharp increase in the number of users of these systems. This, in turn, causes request overload in the systems and reduces system efficiency. Due to the urgency of the problem, numerous studies have been conducted to improve system performance. These methods rely on coding, maintenance, load balancing, and increasing the number of service devices. Unlike existing methods, we propose a request management system for congestion based on priority request servicing. This management system enables full service of conditionally priority requests when all serving devices become busy. A conditional priority badge is assigned to users successfully served by the first stage system. To determine the effectiveness of the proposed request management method, simulation modelling was used due to the cumbersome and complex nature of solving the systems of linear equations. These equations are based on stationary Markov birth-death processes. Assigning a "conditional priority" to successfully processed requests at the initial stage of multi-stage processing and providing preferential service to such requests in the event of overload will increase system efficiency by 5-7%. An algorithm for implementing remote processing of requests with conditional priority is proposed.