Diseño y Desarrollo de un Prototipo de un Cluster en Linux de Alta Disponibilidad para Satisfacer la Demanda de Acceso Web en la Carrera de Ingenierìa en Sistemas Computacionales y el Balanceo de Carga de los Servicios
Fecha
2011-08
Autores
Título de la revista
ISSN de la revista
Título del volumen
Editor
Universidad de Guayaquil. Facultad de Ciencias Matematicas y Fisicas. Carrera de Ingeniería en Sistemas Computacionales. Carrera de Ingeniería en Networking y Telecomunicaciones.
Resumen
El presente proyecto propone una solución al incremento acelerado del internet, el
mismo que se centra en la tecnología de los cluster. La cual permitirán brindar alta
disponibilidad y balanceo de carga de los servicios que proporciona la carrera de
Ingeniería en Sistemas Computacionales a sus usuarios, mediante el uso de
aplicaciones Open Source lo cual representa un ahorro a la institución, es por eso que
se implementará una solución viable que permita soportar la carga transaccional de
peticiones. Dentro de este trabajo se podrá encontrar información relacionada con los
distintos tipos de cluster, características y aplicaciones, seleccionando aquellas que
cumplan los objetivos planteados. El contexto estructural del proyecto se define
mediante un nodo director que gestione y redirija las peticiones hacia los servidores
que tendrán alojados los servicios solicitados. La importancia de los cluster se
fundamenta en la escalabilidad que brindan de poder agregar servidores reales según
las necesidades lo ameriten, al mismo tiempo aumentado el poder de gestión que
representan; además de la redundancia que existen que hacen que trabajen en
paralelo; obteniendo tiempo de respuestas óptimas de rendimiento así como de
puesta en marcha de algún servicio caído. para de esta manera asumir un posible fallo
de uno de los nodos. Permitiendo al sistema funcionar en conjunto donde el usuario
no perciba la caída de algún nodo. Un ejemplo palpable del uso de la tecnología en
mención es google, el cual posee un conglomerado de servidores trabajando como
uno solo, el mismo que atiende miles de peticiones al día; logrando soportar la
demanda diaria de los usuarios. La alta disponibilidad permite agregar, y retirar
servidores sin que comprometan los servidores que formen parte del cluster, además
de no afectar el desempeño total del sistema.
This project proposes a solution to the rapid increase of internet, the same that focuses on cluster technology. Which allow to provide high availability and load balancing services provided by the Engineering in Computer Systems to its users, through the use of Open Source applications which represents a savings to the institution, which is why a solution will be implemented viable to allow the burden transactional requests. In this work, you can find information related to cluster different types, characteristics and applications, selecting those that meet the objectives. The structural context of the project is defined by a node manager to manage and redirect the requests to the servers that have hosted the requested services. The importance of the cluster is based on providing the scalability you can add real servers as required warrant, while increasing the power of management accounting, in addition to the redundancy that exist that make working in parallel, obtaining time yield optimal responses as well as implementation of a service down. thus to assume a possible failure of one of the nodes. Allowing the system to work together where the user does not perceive the fall of some node. A concrete example of using technology in question is Google, which has a cluster of servers working as one, the same that handles thousands of requests a day, achieving withstand the daily needs of users. High availability can add and remove servers without compromise the servers that are part of the cluster, in addition to not affect overall system performance.
This project proposes a solution to the rapid increase of internet, the same that focuses on cluster technology. Which allow to provide high availability and load balancing services provided by the Engineering in Computer Systems to its users, through the use of Open Source applications which represents a savings to the institution, which is why a solution will be implemented viable to allow the burden transactional requests. In this work, you can find information related to cluster different types, characteristics and applications, selecting those that meet the objectives. The structural context of the project is defined by a node manager to manage and redirect the requests to the servers that have hosted the requested services. The importance of the cluster is based on providing the scalability you can add real servers as required warrant, while increasing the power of management accounting, in addition to the redundancy that exist that make working in parallel, obtaining time yield optimal responses as well as implementation of a service down. thus to assume a possible failure of one of the nodes. Allowing the system to work together where the user does not perceive the fall of some node. A concrete example of using technology in question is Google, which has a cluster of servers working as one, the same that handles thousands of requests a day, achieving withstand the daily needs of users. High availability can add and remove servers without compromise the servers that are part of the cluster, in addition to not affect overall system performance.
Descripción
Palabras clave
CLÚSTER, ACCESO WEB