What is the difference between data modelling and simulation modeling in mathematical perspective? By handling complex system and a large scale system of equations for solving a grand challenge application in science and engineering, we describe the mathematical modeling into 2 categories; data modeling and simulation modeling. The analysis of a physical big data required for the accurate analysis and a good prediction. Thus, the concept of data modeling involving a process of designing and developing a data system. A big database is the outcome of an underlying data generation process. Represent the strategy for data flow, the machine learning algorithms for classification and categorization processes are required. In the context of statistical analysis, data-generating process have been presented by a statistical model. The analysis will implicitly adopt some data modeling to describe the performance of the statistical indicators. The data intensive simulation can be supported by high performance computing (HPC) platform based on SIMD or MIMD architecture (Alias, N , 2018). In the context of the numerical analysis, simulation modeling obtained by discretizing the complex mathematical modeling proven by the existence and uniqueness of the solution, theorem, corollary, lemma and proposition properties. The discretization process can be obtained by finite difference method (FDM), finite element method (FEM) or finite volume method (FVM). There are many numerical schemes for solving the FDM, FEM and FVM such as explicit, implicit, semi- explicit, semi-implicit approaches. Grid generation is the most important strategy to generate a big.

Data simulation, to increase the accuracy and to perform the fast solutions. Fine granularity acts as a form of small interval size of domain in space and time step. A large sparse simulation modeling of the discretization method; FDM, FEM of FVM have been supported by huge-memory and high- speed machine types or HPC system (Xing, 2017). This paper will present some grand challenge applications involving complex system model, data intensive modeling and simulation modeling on HPC. The applications deal with medical image processing and EOG signal device. The comparison will be analyzed in terms of the numerical results and parallel performance evaluation for investigating the numerical schemes, and classification strategy. Results, this comparison will benefit to obtain an accurate decision, predictions and trending practice on how to obtain the approximation solution of the data science., conceiving and specifying processes. This Thus, the development of repository software for performing the data and simulation modeling can be expressed as a function of monitoring solution, continuous learning, benchmark designed technology influences end-users to participate and collaborate in different phases of 4 th Industrial revolution (4iR) requirement. As a conclusion, the comparison of data modeling and simulation modeling for intensive data is the alternative framework to fulfill the requirement for generating a fine granular mesh, identifying the root causes of failures and solving a fundamental issue in real time application. Furthermore, the big data-driven and data transfer evolution towards high speed of technology transfer for handling a large and complex system.