首页> 外文会议>International Symposium on Collaborative Technologies and Systems >Collaborative Visualization Spaces for Petascale Simulations
【24h】

Collaborative Visualization Spaces for Petascale Simulations

机译:PetaScale模拟的协作可视化空间

获取原文

摘要

Petascale computing is approaching quickly, but the task of building simulations on these machines is daunting. These computers contain hundreds of thousands of processors, and the simulations can run for weeks to produce vital results, which then must be analyzed and visualized. As the complexity of the simulations and computers increases, so does the complexity of the information that they generate. This data must be carefully analyzed by teams of scientist to achieve the desired scientific insights. In this paper, we focus on one aspect of the Fusion Simulation Project (FSP): collaboration amongst OASCR and OFES. The principal characteristic of the targeted simulations is the coupling of several fusion codes. The principal characteristic of our analytics is that it is highly collaborative in nature. The key to enabling the latter is a backend that allows scientist to perform powerful analyses and visualizations using a web-portal (i.e., our "eSimMon"). Issues such as the orchestration of the flows among simulation, management of storage and analytics resources, movement of the data, and collection of the meta-data are addressed using workflow technologies.
机译:PetaScale Computing正在快速接近,但在这些机器上建立模拟的任务是令人生畏的。这些计算机包含数十万个处理器,并且模拟可以运行数周以产生重要结果,然后必须分析和可视化。随着模拟和计算机的复杂性增加,它们生成的信息的复杂性也会增加。科学家团队必须仔细分析这些数据来实现所需的科学洞察力。在本文中,我们专注于融合仿真项目(FSP)的一个方面:OASCR和OVES之间的协作。目标模拟的主要特性是若干融合码的耦合。我们的分析的主要特征是它在自然界中具有高度协作。启用后者的关键是一个后端,允许科学家使用Web-Portal执行强大的分析和可视化(即,我们的“eSimmon”)。使用工作流程技术解决了模拟,存储和分析资源的流程,存储和分析资源的流程中的流程中的流程,以及元数据的集合等问题。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号