![[ IBM Research ]](masthead10.gif)
| Opal
is a system for very fast scalable distributed real-time interactive visualization
of multidimensional tabular data. Opal is written entirely in java, and
it can operate over the web in a distributed client/server mode to a web
browser applet. Opal can also be conveniently run as an installable non-distributed
stand-alone application.
The Opal architecture is based on a novel compression mechanism for tabular data, for which there is a patent pending. The new data compression mechanism is so efficient that any potential slowdown due to the use of java is easily absorbed. Furthermore, the programs and subroutines which operate upon the compressed data can actually be simpler than programs which might otherwise operate upon the normal uncompressed data. There is a heavily-documented (with javadoc) Opal java toolkit and programming package for building custom new interactive web browser applets using interactive visualization, and also for building custom new servers for the applets. Superceding DiamondOpal is the follow-on to an earlier project called Diamond, which performed similar iteractive multifaceted visualization. Diamond was extremely fast, but it's extensibility, scalability, and distributability were somewhat limited.Opal provides Diamond-like functionality but dramatically relaxes all
of Diamond's constraints, and also dramatically generalizes most of Diamond's
capabilites. Being written in java, it is easy to use and highly extensible.
It can manipulate 10X to 100X more data, and do it at comparable interactive
speeds. And, it is easily distributable in a client/server configuration.
|
||
