An Interactive Visualization Environment for Financial Modeling
on Heterogeneous Computing systems

1. Introduction

Most applications, especially modeling and simulation related applictions,
requires an visualization environment, in which a typical task consists of
three major functional components: a (graphical) user input interface,
a set of computing modules, and a (graphical) user output interface.

3 major components in a visualization environment
user (graphical) input interface ----> a set of computing modules
----> user (graphical) output interface

Usually, in an interactive (animated) visualization environement, the 
input and output interfaces are required to be implemented in certain
graphical forms such that through them the user can interact in real-time
with the computation and the impact of the interaction is immediately
shown in the output interface(usually in display windows). 

The technology of supercomputing provides us with a solution to sovle
a computationaly intensive applications. 
2. Option Price Modeling
. application requirements
. system requirements
. define problem domain

3. AVS

. comprehensive visualization functionalities(graphical programming style, 
  high-level object-oriented design or modular design)
. networking capability(integrator), transparant networking, fault-tolerent, etc.
. highly portability over different vendors (industry standard), XDR

4. Implementation of a interactive (remote) visualization
environment for option price modeling on a hetergenous system

   4.1. Introduction
	We have already had the four option pricing models implemented in
Fortran77 and Fotran90 and run on workstations and CM2/CM5 and DECmpp-12000,
respectively. The main task here is now:
     . design the user input graphical interface
     . design the user output graphical interface
     . distibute the 4 computing modules to machines bset suite to them
     . Integrate the input/output interfaces with the computing modules 
        on different architectures into an unified environment
AVS provides us with both a visualization tool and a high-level networking
tool. Using the AVS library routines, we imlemented portable user input/output
interfaces on SUN, DEC and IBM workstations. Using the remote modules and
network editor, we distributed the interfaces and computing modules on
different machines over Eithernet and integrated them in a single workstation-based visulaization environment. This Section decribes the implementation 
details of this remote visualization environemnt on a hertergenous system.

   4.2. Design of a user graphical interface
	(use screen dumping images as many as possible to show interfaces) 
	. input interface


	. output interface
   4.3. A distributed high performance computing environment
	. distribted computing modules
	. distributed memory usages
	. distributed I/O (and external storages)

5. Discussion
 . a generalized framework using AVS for a interactive remote visualization
   environemt on hertergenous computing systems
 . analysis of system issues: 
	. system integration 
	  hardware: machines from different venders, workstations/supercomputers
          software: multiple programming languages, integration of programming 
	            environemnts on parallel machines(supercomputers) into those 
                    on sequential workstations, networking protocol(software) supports
	            portability, AVS and X-window protocols, the potential of AVS as
	            a system integrator supporting general message passing functionalities
                    in addition to its visualization capability.   
	    
	. system synchronization
	. system portability
 . analysis of visualization issues:
        . 

6. Performance analysis and experiment results

 . derive the general equation and the best case equation for our problem
 . list the analysis timings and the experiment measured timings(speed-ups) for:
     . distributed on SUN, DEC, CM5,DECmpp-12000
     . single machine on DECmpp-12000(AVS on front-machine)
     . single machine on IBM RS6000 (complete sequential code)
 
7. Conclusion