Research And Implementation For Visualization Based On Scientific Workflow On The Grid

Visualization in scientific computing is concerned about generating virtual graphics or images from large-scale data produced by scientific and engineer computing, using computing graphics methods and theories. It has become an indispensable tool in scientific computing, national defense and industry which is inherently so compute and data intensive that it usually needs high-end hardware, massive computing resources and storage resources. Meanwhile, with the development of scientific technology, Internet has become the trend of the world, which is the effective tool to translate messages and cooperate among peoples. The Internet connects thousands of computing resources, storage resources, information resources and etc, which is the important media for production, transmittal and the use of knowledge. And, Grid, as a new computing infrastructure, has covered all these resources geographically-connected and has implemented the function of resource sharing, cooperation and combined computing. It has provided lots of users with all kinds of services in science, engineer, finance, military, etc. The scientific computing has made great progress with the development of Internet. To satisfy the scientists' rapidity need of the scientific computing, the subject's trend is the distributed scientific computing, not only the depth but also width. As a result, the Grid, as an enabling technology to internet wide distributed computing, enables visualization to utilize resources over and provide services to a wider extent of Internet and favors a new research direction, grid-enable visualization.In 1990s, Haber and other people created the model for scientific computing, modeled it as pipelines connected together, which is very important for the development of the scientific computing. Recent years, scientific workflow, as a burgeon noun, is different from business process workflow. Scientific workflow makes procedure execute automatically to achieve the goal according the rules, and the files, messages and tasks can be transformed and executed between the participants. The goal of the scientific workflow is to support convenience for scientists to describe the design of experiments. Right now, the requirements of scientific computing experiments are more and more, which are always reused, modified and shared, so the scientific workflow, as a problem solving environment, is a good tool for scientific computing and a trend for scientific computing. The contents of the thesis are:An extensible three-layer system architecture for scientific computing based on scientific computing on the Grid is proposed, which contains physical layer, logical layer and conceptual layer. Every layer supports independence functions. Physical layer support the foundational functions, which are basement and dynamical uniform visualization service to complete scientific computing. The logical layer binds software resources, parses and explains the scientific visualization computing flow, and manages the scientific visualization tasks. Conceptual layer is the view layer of the system, supporting describing the scientific visualization tasks and interacting with the system users. Low coupling and high extensibility are achieved with this three-layer architecture. Otherwise, this thesis does the performance test, summarizes the result of the results of the test and points the future works.The thesis realizes most of the objects of visualization based on scientific workflow on the grid computing, supporting resource heterogeneity, resource sharing, and task management. The system is platform-cross and extensible, which is good for future works on this research subject.