Load Balancer of Data in a Distributed Network via Nginx Proxy Server

Abstract

Purpose of research. Is investigation of a software method for balancing data in a distributed network via an Nginx proxy server.Methods. In a computer network load balancer of data is an important network parameter. Due to load balancer in the network, the transmission delay may decrease or increase, the spread from the average jitter value. Thus, load balancer in the network affects the time characteristics and network bandwidth. Load balancer can be managed and optimized in both software and hardware ways. The article focuses on load balancer of data at the application level of applications. Hardware load balancer, which is solved within the framework of network equipment itself, for example, in switches, is briefly considered. This is handled by the queue manager in the Ethernet switch, which manages the bandwidth and queues. Cyclic algorithms are described, as well as an algorithm with time selection of frames in dispatcher of switch that implement effective hardware load balancer. Software load balancer of data in the network is considered. A web server and an Nginx reverse proxy server were used as software load balancer, 3 Docker containers based on Asp.net applications running on different environments.Results. The network was configured and the cyclic load balancer algorithm was used in the Nginx server. A research of a network with a different number of environments in the network, web servers, data requests was conducted. The cyclic load balancer of data in Nginx is more efficient than the random algorithm, this has been shown during experiments.Conclusion. Hardware and software load balancer algorithms in a distributed network were considered and investigated. Cyclic load balancer of data has made it possible to increase the network bandwidth, its efficiency and performance.