viewpoints   high-performance interactive linked scatterplots and more

Overview

---

Viewpoints (vp) is a high-performance visualization and analysis tool for large, complex, multidimensional data sets. It allows interactive exploration of data in 100 or more dimensions with sample counts, or the number of points, exceeding 10⁶ (up to 10⁸ depending on available RAM). Viewpoints was originaly created for use with the extremely large data sets produced by current and future NASA space science missions, but it has been used for a wide variety of diverse applications ranging from aeronautical engineering, quantum chemistry, and computational fluid dynamics to virology, computational finance, and aviation safety.

Viewpoints does a modest set of useful things fairly well (or, at least it does them rather quickly). It supports multiple linked scatterplots (millions of points in 2D & 3D) with overplotting and brushing. It supports histograms with dynamically adjustable binwidth. It supports several common types of normalization (linear, logarithmic, logistic, trimmed, etc), equalization (gaussian, uniform, etc) and rudimentary outlier detection and forward search.

Viewpoints is designed to take advantage of the hardware accelerated graphics and fast processors on modern high-end workstations and laptops. It is written entirely in C++. It uses OpenGL for graphics, gsl and gmm++ for numerics, boost, and blitz++ for data management, and fltk for the graphical user interface. It is highly interactive, so that most parameters are modifiable in real time with immediate visual feedback. It stands apart from most "multiple linked views" graphics applications because of its high performance on large data and its availablility on multiple platforms.

Viewpoints is inspired by the hyperwall, a 50 node linux cluster with a 7x7 array of tiled displays, developed at NASA Ames Research Center specifically for large scale high-dimensional data visualization. The hyperwall has been sucessfully applied to problems in astrophysics, earth science, aerodynamics, and life sciences. But unlike the hyperwall, viewpoints does not require custom hardware or a dedicated visualization facility. Instead, it is designed to live gracefully on any working scientist's desk (or lap) and will allow them to quickly explore and visually analyze the most common forms of large multi-dimensional scientific data.

Creon Levit wrote the initial version of viewpoints for OSX and Linux in 2005. It is currently under active development by Paul Gazis and Creon Levit at NASA Ames Research Center.

Viewpoints is funded by NASA's highly successful Applied Information Systems Research program (AISR). The purpose of the AISR program is to exploit advances in information science and technology to enhance space science productivity from NASA's space flight missions. For more information about the AISR program, see http://aisrp.nasa.gov.

Application binaries are available for Linux, Apple OSX (PowerPC and Intel macs), and Microsoft Windows. An interface to MATLAB is also available. For more information, and/or to obtain a copy of the latest beta release, see below.

Because viewpoints is a development system, we would appreciate it if you could provide us with user information so that we can track usage to gain a better understanding of the needs of the user community. This information will remain private, and will not be distributed in any way!