The supported grid tool for measuring an applicability of medical dataset visualization

Abstract

Grid computing is the technology that allow scientist to share their personal computer to store, process and visualize a large amount of data economically and efficiently. Medical dataset such as Computerized Tomography (CT) Scan and Magnetic Resonance Imaging (MRI) may contain a huge size of data that requires a powerful computer to visualize it. Existing grid tools and middleware are used to monitor and manage grid resources to achieve high computing processor. However, validating the medical dataset and measuring appropriate number of resources is an important task that may reflect the overall performance of grid computing. For example any damages in a dataset or missing in any sequence slices may result an error in processing. Additionally the number of resources for a specific size of dataset is essential to be identified. This project is aimed to study existing grid performance tools and develop the supported grid tool to measure an applicability of medical dataset in visualization process. Problem formulation and scope identification is the first step taken. Some analysis on grid performance tools namely Ganglia, Hawkeye and GridIce is performed to identify the performance criteria and supported grid tool development and evaluation is finally obtained. The developed supported tools is using dataset scanning technique and be able to identify the suitable and non-suitable medical dataset for visualization. The tool shows that the requirement of medical dataset with the size of 60 MB is six standard grid resources.