Design and Implementation the High-Performance PC Cluster Architecture with Diskless Slave Nodes in the Computer Classroom

Abstract

The objective of this thesis is to design and implement a highperformance computing environment by clustering idle desktops with diskless slave nodes on campuses. Those PC Clusters can be easily switched from “general operation” to “high-performance computing” and vice versa. It is demonstrated by simulation results of the matrix multiply with LAM/MPI and PVM. The bioinformatic program, FASTA, is executed smoothly in the Cluster architecture as well. In addition, the system and network monitoring software, ntop and ganglia, are installed in the PC Clusters for the system performance tuning in the future.