Abstract:
On top of the need for high performance, high computing power, resource sharing and virtualization challenges, this research is intended to deal with grid and cluster computing technologies to Support high availability solution, and prototype an implementation of tightly coupled cluster web services, testing the availability of web services incase of node failure.
This project aims to apply the concept of high availability on the web page using grid cluster computing technology, and defines grid computing and cluster computing from the perspective of the Linux operating systems through the use of virtual IP. Which distribute the load on the active servers that contain the page within the clusters node when the nodes are tightly coupled and loosely coupled. It gives a clear understanding of cluster computing and grid computing options, topologies, layers, and how the two technologies complement each other.
The testbed constructed to implement high availability solution on both tightly and loosely coupled web services, and take a deep analysis of cluster configuration tools and services, furthermore pointing out some results when using deferent types of failure, network failure or software failure, then compare the performance and efficiency.
