Overview
Management Team
Fact Sheet
Quotes
Press Releases
Events
Collateral
Overview
Newsletters
Image Library
Partners
Developer's Corner
Employment Opportunities


White Papers

  • JWAVE Architectures
    How JWAVE can provide solutions in a wide variety of application architectures
  • Using JWAVE 3.5 in Server Side Java Applications
    The move from Applets to server-side J2EE applications JWAVE is designed and presented as a solution for client/server and custom web-architected solutions. It has broad functionality and an open design that allows it to fit into many architectures and applications. Through the use of a Java servlet, the JWaveJSPServlet, it can also fit into server-side J2EE applications where no client-side Java is used. It can be used for either custom stand-alone solutions, or solutions involving integration with other tools including Application Servers, and N-tier architectures.
  • Calling the IMSL C Numerical Library from Java
    This report describes how to use the Java Native Interface (JNI) to call an IMSL C Numerical Library (CNL) function from Java. Using JNI, Java calls C interface code, which then calls the CNL function.
  • JWAVE and CORBA
    JWAVE is a distributed data analysis and visualization solution, using PV-WAVE as a server engine to provide the horsepower for analysis and visualization tasks. As JWAVE is a multi-platform, networked, distributed solution, it is natural to ask the question, "Can JWAVE plug into CORBA?"
  • The Problem with Portals: Envisioning the Best Technologies for the Enterprise Information Portal
    The Enterprise Information Portal (EIP) is a simple concept and an effective solution to an age-old problem. Properly implemented, it can transform knowledge workers¡¦ PCs into a single doorway to enterprise information wherever it may happen to reside. The potential benefits of EIPs are straightforward enough. The problem with portals, however, is that just providing knowledge workers with a doorway to enterprise information is not the same as providing them with the insights they need to identify and seize business opportunities or to spot and address potential problems or threats. It¡¦s not just information that knowledge workers need; it¡¦s timely insights that spur action.
  • VDA Tools Technology Unleashed
    For many PV-WAVE programmers, their exposure to VDA tools consists of running the Navigator and using it as an interactive point-and-click method of accessing PV-WAVE functionality. Few users actually program using VDA tools or the underlying technology. The reason for this is certainly the great complexity of the VDA tools Visual Numerics provides and the great number of underlying routines they make. The goal of this document is to try and uncover some of this complexity and show how both the current VDA tools and the underlying pieces of the VDA tool technology can be easily applied in new applications and retrofit to existing applications.
  • Sensor Response Modeling for Trackers
    The primary objective of a sensor response model is to provide object sighting messages (OSMs) in the form of accurate position and amplitude estimates to the tracker. This paper will examine the capabilities of a sensor response model entitled ¡§Passive Sensor Workbench〃 to evaluate object sighting measurement accuracy by implementing a selected candidate sensor design and signal/image processing technique.
  • Integrating Simulation Research Into Curriculum Modules
    Visual Numerics describes the development of modules for teaching a senior-level course, Mechanical Behavior of Materials, that incorporate the results of state-of-the-art simulation techniques. The modules are Java Web-based. These modules teach the basics of materials mechanical behavior using research simulation codes that are state-of-the-art in the materials simulation community.
  • Enterprise Computing for Numerics and Visualization
    The way that people use computers is in a state of turmoil. Desktop-based computing is moving to a world in which the enterprise becomes more important than any one individual machine: the enterprise computing model. In this white paper, Visual Numerics shares its vision of how worldwide web technology and distributed computing are converging to make enterprise computing technical computing a reality. As a computing professional, the technology roadmap for enterprise computing numerics and visualization that Visual Numerics outlines in this white paper is vital for your own technical computing and application strategy.
  • JWAVE in the CAVE at Virginia Tech
    In partnership with the National Computational Science Alliance (NCSA), Virginia Tech's Visualization Laboratory has taken the next giant step in scientific visual data analysis of volumetric, three-dimensional data - by building the CAVE (VT-CAVE). Concurrently, Sun Microsystems, Inc. and Visual Numerics, Inc. co-sponsored the creation of the Scientific Modeling and Visualization Classroom (SMVC). Together, the VT-CAVE and SMVC allow students and faculty link to high-performance computers and visualize their results at their desks or in the CAVE. This article will describe how this link is being built for educational and research projects, and will focus on how Java™ and JWAVE was used to model, simulate and visualize results.
  • Evaluating American-Style Call Option Prices Based on Constrained Least-Squares
    Computing the current value of an American call option requires solving the Black-Scholes PDE. The numerical computation involves solving this free boundary problem. A finite difference approximation leads to a linear complementarity problem with a tridiagonal matrix. With this paper Visual Numerics examines an efficient technique for solving a related quadratic programming problem. This development is based on a non-negative constrained least-squares (NNLS) algorithm. Three examples are given, including graphics illustrating the solutions.
  • Creating DLLs for Use by LINKNLOAD in PV-WAVE® on Windows
    This document describes the steps to create a DLL for use by LINKNLOAD in PV-WAVE on the Windows platform using Microsoft Developer Studio 97 with Visual C++ 5.0. This document is meant to supplement the material found in the LINKNLOAD entry in the on-line help. Although the emphasis of this document is on using LINKNLOAD on the Windows platform, LINKNLOAD users on other platforms may find some of the examples beneficial.
  • Aasen's Method for Linear Systems with Self-Adjoint Matrices
    Aasen's algorithm is effective for solving systems of linear algebraic equations with self-adjoint matrices. This note gives the formulas for complex matrices and also mentions a technique for stabilizing the diagonal elements of the tri-diagonal factor.
  • High Performance Fortran Versus Explicit Message Passing on the IBM SP-2
    The goal of this work was to provide high-performance parallel implementations of the LU, QR, and Cholesky factorizations and their associated solvers in double precision and double precision complex for the IBM SP-2 for Visual Numerics. For ease of development and maintenance, the hope was that this could be done by implementing these routines in High Performance Fortran (HPF), rather than using explicit message passing. For usability with existing VNI libraries, these implementations were required to work with advanced Fortran 90 features such as derived data types, modules, user-defined operators, and user-defined generic procedures.
  • Task Allocation and Performance Measures for the Box Data Type
    The Visual Numerics, Inc. IMSL Distributed Network Fortran Library product has generic functions and operators. One of the types valid for these functions is the box data type. This paper will illustrate the basic algorithms and design for the DNFL box data type functions, outline the basic task allocation algorithm, present timing data that shows typical performance measurements, and present results for matrix inverses, solutions of linear algebraic equations, matrix products, Cholesky factorizations and two-dimensional discrete Fourier Transforms.
  • VDA Tool Technology
    This white paper gives an overview of the exciting new features of PV-WAVE. The VDA Tool Architecture is a framework for developing cross-platform applications with graphic-user interfaces (GUIs) in PV-WAVE. It is intended to put the VDA Tool Architecture into context and position the PV-WAVE Environment as a development tool.
  • Connecting PV-WAVE to ArcView and ARC/INFO
    Many people see the advantage of using PV-WAVE from Visual Numerics with ARC/INFO from ESRI. Each corporation has a number of products that capture the worldwide market share of their respective markets. Both product lines support an open and flexible environment where world-class solutions can be developed using the outstanding features of both products. This paper describes the available connection methods.
  • Using PV-WAVE as a Web Browser Helper Application
    The goal of a helper application is to allow users to distribute PV-WAVE .pro and .cpr files over the Web. Click on a URL, which links to a Wave file, and Wave fires up and executes the application on your machine.
  • PV-WAVE Net Possibilities
    The possibilities are endless. This paper includes just a few of the ideas with links to examples. The basic methods are: Offline Picture Generation, Server Applications - Batch Mode, Server Applications - Interactive Mode, Client Plug-In, and Client Helper Applications.
  • Efficient Programming with PV-WAVE Widgets
    This paper details the recommended programming practices within PV-WAVE Widgets to reduce the occurrence of errors generated due to the lack of virtual memory. The intended audience is all user's of PV-WAVE Widgets concerned with memory management and the use of temporary variables in PV-WAVE Widgets. The core purpose of this document is to enlighten the software developer as to a more efficient ways of making use of PV-WAVE Widgets so that their applications may grow to be more robust and usable.



¤½¥q简¤¶ 产«~ÉOªA务 ¸Ñ决¤è®× ¦¨¥\®×¨Ò §Þ术¤ä«ù ¤å¥ó¤U载 传°e¥»页给±zªºªB¤Í
© Copyright 2008 Visual Numerics, Inc. ¬ü°Ó«Â¯à«H®§ ª©权©Ò¦³ µÛ§@权声©ú 隐¨p声©ú