Analysis and Implementation of a Parallel Computing Cluster for Solving Computational Problems in Data Analytics
Chavala Mutyala Rao, K. Shyamala · 2020 5th International Conference on Computing, Communication and Security (ICCCS) · 2020
As computers get updated by a couple of generations, they get disposed of as useless but when a large number of such computers are put together and they share each other's computational power, they can get really powerfulcluster defines as a parallel computing system which is inter connected with a group of independent computers working togeather as a single coherent system. Clusters uses message passing interface for communication, and code snippets have to be explicitly programmed to run on underlying hardware. More computing nodes to be added to the cluster based on the need. While designing parallel cluster in many cases we opt for open source technologies and Linux system components which reduce software costs. Parallel Cluster is a way of connecting similar computers together and can get a task done by sharing the task load. The task is shared in coordination of the master and slave nodes in a cluster. Parallel computing clusters allows group of computers with different hardware configurations, to work together to provide solutions to different computational problems. The resources like memory, storage and CPU cores in a cluster computing environment are integrated into one system. This paper is to set up a cluster environment for academic purposes and provide a huge computation power to run a program in parallel mode this will reduce the time taken to execute a program. A parallel code is always better then sequential code in terms of time taken to run.