Journals Proceedings

Abstract

Energy consumption for high performance computing is enormous. This can be substantially reduced using Cloud Computing which provides low cost, energy efficient and ubiquitous computing. But Cloud Computing cannot work efficiently with limited resources. We aim to build a Private Small Cloud (PSC) that is based on small clusters, virtualiz ation and graphics processor which can handle large amounts of data with low cost, high efficiency and reliability using limited resources. Using Cloud Computing for low power PC cluster architecture and low power client devices can reduce energy consumption and carbon emission. Virtual machine server run multiple operating systems simultaneously on a single computer and supports multiple heterogeneous applications running on a number of client devices. General purpose graphics processing unit (GPGPU) makes use of parallel processing and multithreading. The cloud is designed to work with Linux thin clients using wired connection.