A Framework for Online Visualization and Simulation of Critical Weather
Preeti Malakar, Vijay Natarajan & Sathish S Vadhiyar
Critical weather applications like cyclone tracking and earthquake modeling require
high-performance simulations and online visualization simultaneously performed with
the simulations for timely collaborative analysis by geographically dist
ributed climate science community. A computational steering framework for contro
lling the high-performance simulations of critical weather events needs to take
into account both the steering inputs of the scientists and the criticality need
s of the application including minimum progress rate of simulations and continuo
us visualization of significant events. In this work, we have developed an
integrated user-driven and automated steering framework for simulations, online
remote visualization, and analysis for critical weather applications. Our framework
provides the user control over various application parameters including region of
interest, resolution of simulations, and frequency of data for visualization.
However, the framework considers the criticality of the application, namely, the
minimum progress rate needed for the application, and various resource constraints
including storage space, network bandwidth, and number of processors, to decide on
the final parameter values for simulations and visualization. Thus our framework
tries to find the best possible parameter values based on both the user input and
resource constraints. We have demonstrated our framework using a cross-continent
steering of a cyclone tracking application involving a maximum of 128 processors.
Experimental results show that our framework provides high rate of simulations,
continuous visualizations, and also performs reconciliation between algorithm and
user-driven computational steering.