GPU-Accelerated Web Application for Metocean Descriptive Statistics

In today’s real world application, many researchers and developers are using Graphic Processing Unit (GPU) to accelerate non-graphical application. Modern GPUs which are massively parallel general-purpose processors has a big advantages on big data analytics in terms of power efficiency, compute density and scalability. In oil and gas industries, metocean data is being generated, collected and analyzed at an unprecedented scale. Metocean data which is observed measurements of current, wave, sea level and meterological data are regularly collected by major oil and gas companies. This data, is usually collected by specialist companies and distributed to paying parties who will deploy their scientist and engineers to analyze and forecast information based on the information. The analyses of metocean data provide crucial information needed for operation or design work that has health, safety and environment (HSE) and economic consequences. Therefore, this paper proposed GPU-accelerated web applications for metocean descriptive statistics to improve the current CPU based implementation. Metocean descriptive statistics is the analysis of metocean data that helps provide important information required for operation that has health and safety as well as economic consequences. The application will utilize GPU to perform descriptive statistics for metocean data. The implementations of GPU for metocean descriptive statistics is expected to provide a better raw performance, better cost-performance ratios, and better energy performance ratios. The main objective of this project is to develop a GPU-accelerated application for metocean descriptive statistic and a web-based application that links to the GPU-accelerated application, besides demonstrate the capabilities of GPU in performing non-graphical calculation