Alta disponibilidade e balanceamento de carga para servidores web usando linux

Abstract

This paper aims at implementation of a cluster for Web services that have high availability and load balancing in case of failures. Clusters of high-availability systems are designed to provide availability of services on an ongoing basis, using redundant computers. If a computer fails, another automatically takes over running the same services and data. Failures of services are due to hardware problems or overload of services on servers. The problem of overloading could be soften using load balancing, is distributing the requests / traffic with other computers in the system, running the same programs, reducing the processing time and improving the performance of the application. We evaluated the methods of implementation of clusters, their models and concepts which the best results to ensure a quality level compatible with the proprietary solutions available. It is shown that the use of cluster is effective in balancing and high availability.