Integration of a Big Data Emerging on Nanotechnology Theory, Model,
Simulation and Its Application using Green Computing Platform
By Norma Alias
Center for Sustainable Nanomaterials,
Ibnu Sina Institute for Scientific and Industrial Research, Universiti Teknologi Malaysia
A small data solution dealing with a simple analysis, reporting, course-grained visualization with giga-bytes or a few tera-byte data sizes. A single processor and small size of the local memory platform are well be enough to support small scale simulation. Contradictory, a big data integrated solution was not just limited to analyzing and reporting, but high potential for predicting, insights with complex, advance and fine-grained resolution. Multi-structured data type size becomes a significant consideration of peta-, exa- and zetta-bytes data size. Thus, small data size is not well suited for predicting and solving nanotechnology application accurately. A big data associated with nanoscale computing can be generated by sensor data, device data, image processing resolution and grid generation simulation. These emerging opportunities to combine big data integration with nanotechnology theory, multi-scale model and large scale simulation. Furthermore, the nanotechnology theory includes the fundamental nanoscale structures, features and mechanical properties of the nanomaterial. The multi-scale model involving the integration of ordinary differential equation (ODE) and partial differential equation (PDE) with multi-type system and multi-scale features. The discretization techniques for solving the multi-scale mathematical model emphasized on mesh generation strategy of finite difference method (FEM) and finite element method (FEM). Green computing of high performance platform has been used to support the big data environment and large sparse simulation. Green computing includes the implementation of energy-efficient of sustainable IT facilities such as CPUs, GPU, servers and peripherals as well as reducing resource consumption and proper recycle of low cost computer equipments.
Additionally, this paper proposes some conceptual frameworks for big data integration on five nanotechnology applications. First application is the nanoparticle assisted drug delivery process through a blood flow. Second application is the fabrication of silicon nanowire by chemical vapor deposition (CVD) process. Third applicationis the prediction of some depend parameters on multilayer nanoscale device in semiconductor manufacturing. Fourth applicationis molecular abnormal cell growth and the fifth is an image processing for tumor cell classification. Parallel algorithm for the numerical solution emphases on SIMD taxonomy instructions. In order to improve the performance on green computing system, this paper investigates the distributed-shared memory architecture containing multi-core Intel Xeon processors and CPU-GPU platforms. Numerical analysis and parallel performance evaluation (PPE) are the indicators to validate the nanotechnology complex theory, multi-scale model and large scale simulation of the grand challenge applications. Comparison table, graph illustration and 2D visualization are the tools for result presentation.
Short Bio A.P Dr. Norma Alias
A.P Dr. Norma Alias is Research Fellow at Center for Sustainable Nanomaterials, Ibnu Sina Institute for Scientific and Industrial Research, Technology University of Malaysia. Currently, she is also Associate Professor at Mathematics Science Department, Faculty of Science, Technology University of Malaysia. She received her M.S. degree and PhD from National University of Malaysia, in Computer Science and Industrial Computing. Area of research interest includes mathematical modeling and simulation, big data computing, large scale image processing, and computational nanotechnology on high performance computing system. She has published more than 160 research papers in ISI, SCOPUS, index and non-index journals. She won 9 international and local innovation medals. She has awarded as 40 active researchers in Malaysia and completed 30 research grants.