高级检索
    徐平均 沈卫超 廖 丽. JaVis系统中的多分辨数据组织与交互可视化[J]. 计算机研究与发展, 2010, 47(6): 996-1004.
    引用本文: 徐平均 沈卫超 廖 丽. JaVis系统中的多分辨数据组织与交互可视化[J]. 计算机研究与发展, 2010, 47(6): 996-1004.
    Xu Pingjun, Shen Weichao, and Liao Li. Multiresolution Data Organization and Interactive Visualization in JaVis[J]. Journal of Computer Research and Development, 2010, 47(6): 996-1004.
    Citation: Xu Pingjun, Shen Weichao, and Liao Li. Multiresolution Data Organization and Interactive Visualization in JaVis[J]. Journal of Computer Research and Development, 2010, 47(6): 996-1004.

    JaVis系统中的多分辨数据组织与交互可视化

    Multiresolution Data Organization and Interactive Visualization in JaVis

    • 摘要: 随着高性能计算机以及相关软硬件技术的飞速发展,数值模拟的规模越来越大,置信度越来越高,其产生的数据场规模亦越来越大、越来越复杂,需要更高级的科学计算可视化方法分析这些数值计算结果.然而,针对目前的TB级以及更大规模数据场,可视化的交互性能迟滞落后,成为阻碍数据分析的重要因素.如何提高可视化分析的交互速度,减少数据场的I/O处理是目前可视化系统必须解决的重要问题.多分辨技术是解决大规模数据场交互可视化的重要手段之一.自研发的大规模并行分布式数据分析与可视化系统JaVis采用多分辨技术提高数据可视化的交互速度,其中实现的关键技术如下:多分辨数据的组织、多分辨控制插件的生成、多分辨率层次切换技术等,并利用物理科学数据进行了性能及可靠性测试.

       

      Abstract: With the rapid development of high performance computer and technology, scientists can model complex physical phenomena at a scale and fidelity to answer some of the most important and difficult questions in science. The output from these simulations, however, is so voluminous and complex that advanced visualization technologies are necessary to interpret the calculated results. Even though visualization technology has progressed significantly in recent years, it is difficult to visualize and analyze terascale or much larger data interactively to full extent. Interactive speed of dealing with large data embarrasses the operation of data visualization and analysis severely. How to reduce I/O data size and improve interactive speed are the main challenges in developing visualization system designed for large scale data analysis. Many studies indicate that multiresolution technique is an important and feasible means of addressing these problems. In this paper the authors present some new multiresolution techniques implemented in JaVis, a large-scale data analysis and visualization system, to accelerate interactive visualization of large dataset. The key techniques developed are summarized as follows: multiresolution data organization, multiresolution data operator plugin implementation and multi-level data switch. Also demonstrated are the efficiency and robustness of the interactive visualization techniques with some real physical datasets.

       

    /

    返回文章
    返回