Steering and Visualization of Electro-Magnetic Simulations Using the Globus Implementation of a Computational Grid
نویسنده
چکیده
A framework for computational steering of a finite difference code for electromagnetic simulation has been developed and implemented. In computational steering we need to develop software which allows the user to enter an interactive visualization or VR environment and from there control the computation. A proof of concept implementation has been carried out using an existing code for 3D finite difference time domain approximation of Maxwell’s equations. Large parts of the computational steering software are general but details in the choice of control variables and visualization is specialized to the electromagnetics code. To handle the large computational requirements of both simulation and visualization the system can be distributed across multiple machines. This is possible through the use of the Globus toolkit for communication, data handling, and resource coallocation. It program also makes use of VTK for data filtering and the generation of visualization elements, and IRIS Performer with pfCAVELib for 3D interactive rendering on CAVE compatible devices. Two testcases are presented. In one example with a smaller number of computational cells, full computational steering with recomputation is possible. In another with a large number of computational cells the solution is precomputed and only the visualization is interactive. The scalability of the computational code is tested for different computers in order to determine the size of the problem which can be handled with full computational steering on the available local hardware.
منابع مشابه
Computational Steering Systems in Grid Computing Environments
Computational steering is a powerful idea to enable scientists to interactively control running simulation code. With the emergence of Grid technologies and e-Science, experts are expected to collaboratively steer more complex and geographically distributed simulations. This paper investigates several representative Grid-based computation steering systems. These systems are used to create colla...
متن کاملThe TeraGyroid project – collaborative steering and visualization in an HPC Grid for modelling complex fluids
The TeraGyroid experiment[1] addressed a large scale problem of genuine scientific interest and showed how intercontinental Grids enable new paradigms for collaborative computational science that can dramatically reduce the time to insight. TeraGyroid used computational steering over a Grid to study the self-assembly and dynamics of gyroid mesophases (found in novel materials and living cells) ...
متن کاملPractical Tools for Computational Steering
Computational steering refers to the real-time interaction of a scientist with their running simulation code. Despite the many benefits associated with computational steering, its uptake to date has been limited. In this paper we discuss the reasons for this and how the computational steering library and associated tools developed as part of the RealityGrid project aim to tackle them. We descri...
متن کاملgViz: Visualization and Computational Steering on the Grid
Visualization and computational steering are fundamental aspects of Grid computing. Visualization is important in the analysis and interpretation of the results generated by large scale simulation; computational steering enables effective use to be made of interactive access to high performance computing. This paper describes work in the gViz e-Science project: a reference model based on the da...
متن کاملCOLLABORATIVE INTERACTIVITY IN PARALLEL HPC APPLICATIONS Interactive Computational Steering of Grid Applications
Large-scale scientific research often relies on the collaborative use of massive computational power, fast networks, and large storage capacities provided by e-science infrastructures (e.g. deisa, egee, etc.) since the past several years. Especially within e-science infrastructures driven by high-performance computing (hpc) such as deisa, collaborative online visualization and computational ste...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2000