Cluster Computing: High-Performance, High-Availability, and High-Throughput Processing on a Network of Computers

This paper discusses the motivation for cluster computing as well as the technologies available for building cluster systems. The emphasis placed for using commodity-based hardware and software components to achieve high performance, availability, and scalability means that cluster computing is a more cost-effective platform compared to traditional high performance platforms. It examines the various cluster-specific components such as interconnection technology, operating system, middleware and programming model. The paper discusses the performance of a number of common cluster interconnects. It also presents various parallel programming models and concepts of single system image and its realization at the cluster resource management level.