@ARTICLE{pumasteering,
   author = "Modi, A. and Sezer, N. and Long, L. N. and Plassmann, P. E.",
   title = "{Scalable Computational Steering System for Visualization of Large S
cale CFD Simulations}",
   howpublished = "32nd AIAA Fluid Dynamics Conference and Exhibit, St. Louis, M
issouri",
   journal = "AIAA 2002-2750",
   month = Jun,
   year = 2002
}

Old Title: Solution and Visualization of Complex Flow Simulations using an Interactive Computational Fluid Dynamics System
New Title: Scalable Computational Steering System for Visualization of Large Scale CFD Simulations
Authors: Anirudh Modi <anirudh-modi@psu.edu>, 
	Nilay Sezer-Uzol <nxs216@psu.edu>, Lyle N. Long <lnl@psu.edu>,
	and Paul E. Plassmann <plassman@cse.psu.edu>
        Institute for High Performance Computing Applications (IHPCA)
        Pennsylvania State University
Presenter's Name: Anirudh Modi
Presenter's email: anirudh-modi@psu.edu
Presenter's phone: (814) 865-196

Parallel simulations are playing an increasingly important role in
all areas of science and engineering. As the applications for these
simulations expand, the demand for their flexibility and utility grows.
Interactive computational steering is one way to increase the utility of
these high-performance simulations, as they facilitate the process of
scientific discovery by allowing the scientists to interact with their
data. On yet another front, the rapidly increasing power of computers
and hardware rendering systems has motivated the creation of visually
rich and perceptually realistic virtual environment (VE) applications.
The combination of the two provides one of the most realistic and
powerful simulation tools available to the scientific community. While
tremendous amount of work has gone in developing Computational Fluid
Dynamics (CFD) software, little has been done to develop computational
steering tools that can be integrated with such CFD software. Here at
Penn State, we have developed an extremely simple to use general-purpose
computational steering library which can be coupled effortlessly to any
C/C++ simulation code. The library is written in C++, using several
of C++'s advanced object-oriented features, making it very easy and
powerful to use while hiding most of the complexities from the user.
The library allows a simulation running on any parallel or serial
computer to be monitored and steered remotely from any machine on the
network using a simple cross-platform client utility. We have used this
library to augment our parallel flow solver, Parallel Unstructured
Maritime Aerodynamics (PUMA), which is written in C using the Message
Passing Interface (MPI) library, to obtain a powerful interactive CFD
system. This system is being successfully used to monitor and steer
several large flow simulations over helicopter and ship geometries, thus
providing the user with a fast and simple debugging mechanism, where the flow
and convergence parameters can be changed dynamically without having to
kill or restart the simulation. This CFD system, which primarily runs
on our in-house Beowulf Cluster, the COst-effective COmputing Array
(COCOA), has been coupled to our Virtual Reality (VR) setup to obtain
near real-time visualization of the 3D solution data in stereoscopic
mode. This ability to get "immersed" in the complex flow solution as
it unfolds using the depth cue of the stereoscopic display and the
real-time nature of the computational steering system opens a whole
new dimension to the computational fluid-dynamicists for interacting
with their simulations.