Analysis of High Performance Computing in Computer Resource Sharing for Parallel Processing Using OpenMPI
Sujoko Sumaryono, Freddy Kurnia Wijaya, Teguh Bharata Adji · 2011
High performance computing can be associated with a method to improve the performance of an application that has a large computational process. This includes the distribution of computer resources such as processing load, CPU usage and memory usage into multiple units that enable simultaneous distributed processing to improve speed in problem solving. Cluster system for parallel programming using message passing algorithm interface (MPI) with the implementation of openMPI to test and analyze the ability to distribute the computing process to several computer resources i.e. the processor load, CPU usage, and memory usage as well as the association with network resources i.e. traffic, throughput, and packet. The method used is descriptive analysis by observing responses of the system designed. Performance analysis was conducted by running parallel programs of which processes were executed at some computer nodes. The relationship between computer resources and network resource is monitored by using ganglia application and the network resource is monitored by using ntop. This research shows that the network and computer resources result in a better computation time compared to sequential computation time.