From InfovisWiki
What is Titan? The Titan Informatics Toolkit is a collaborative effort between Sandia National Laboratories and Kitware Inc. It represents a significant expansion of the Visualization ToolKit (VTK) to support the ingestion, processing, and display of informatics data. By leveraging the VTK engine, Titan provides a flexible, component based, pipeline architecture for the integration and deployment of algorithms in the fields of intelligence, semantic graph and information analysis. Why Titan? The Titan project represents one of the first software development efforts to address the merging of scientific visualization and information visualization on a substantive level. The VTK parallel client-server layer will provide an excellent framework for doing scalable analysis on distributed memory platforms. The benefits of combining the two fields are already reaping rewards in the form of functionality such as the cell lineage application below. How do I use it? In the same way that scientific visualization applications can be built with VTK, you can now build information visualization and analysis applications with Titan. As shown in the image to the right, applications can be built by combining the components as appropriate for your specific domain needs. If you're interested in using Titan for development the Titan developers guide is the right place to start
- Titan Developers Guide
|
An application based on the Titan Informatics Toolkit is shown here displaying multiple views including table, graph and geospatial all semantically linked.
|
Documentation
|
|
|
General Titan information
-
How to obtain the source and build Titan.
-
Problems/solutions to installing Titan apps on Windows.
-
Other information visualization toolkits.
Technical documents
-
How to add graphviz graph diagrams to the wiki.
-
How to link to other wikis and bug trackers from this wiki.
-
Migrating Titan classes into VTK.
-
Just some administrative blah...
Algorithms
-
How to create a new algorithm in Titan.
-
How to create a new reader in Titan.
-
How the G-Space graph layout strategy works.
Technical documents
-
Proposal
-
In progress.
-
Multicore toolkits that Titan could integrate with.
ChartsTechnical documents
-
Design discussion about the new Qt charts. Completed.
-
How charts currently work in ParaView.
-
Discussion on how we could get charts integrated into OverView/ParaView.
-
Final decision on how to get charts integrated into OverView/ParaView. In progress.
-
Work on even newer charts using Cairo. In progress.
Code integration
-
Scripting Titan using Python.
-
Write Matlab code that uses Titan components.
-
Write Titan components that use existing Matlab code.
-
Integrating Titan with Java.
-
Integrating Titan with the R statistics language.
-
Guide for building and running R and Matlab with VTK for common computer platforms
|
Databases
-
How to read data from a database into a Titan graph.
-
Another explanation of how to convert tables to graphs. These two pages need to be merged.
-
Description of basic methods available in database classes.
-
Table of database types to Qt and VTK basic types.
Technical documents
-
Research on how we could convert database structures to ontologies.
-
Discussion on the current version of the graph and tree data structures. Completed.
-
Timing of vtkGraph and Boost Graph Library data structures.
Data structures
-
Description of new vtkSelection data structure.
Technical documents
-
Discussion of how to implement large binary objects. Proposal.
-
Caching results of operations by saving to a database. Proposal.
-
Research on how the PBGL data structure works.
-
Discussion on how to integrate PBGL and vtkGraph. Proposal 2 is the currently implemented structure.
-
Implementing the new attribute type "Pedigree IDs". Completed.
-
Supporting Unicode strings in VTK. In progress.
-
Discusses mechanisms to simplify downcasting from vtkArray to vtkTypedArray<T>.
-
Current work on file I/O for the N-Way arrays.
-
Random thoughts on how to handle distributed dense and sparse arrays.
Geospatial visualization
Technical documents
-
How to dynamically prune the image tree. Completed.
-
Ideas on how geovis can be integrated into OverView.
-
Description of the old threading geovis worker thead models before they were integrated.
Linear AlgebraTechnical documents
-
Research on various linear algebra packages that could possibly integrate with Titan.
|
Statistics
-
Titan offers a set of statistics engines, some of which already have a scalable, parallel implementation.
Text Analysis
Views and rendering
-
The state of the art of labels in VTK.
Technical documents
-
Design for annotations in VTK. In progress.
-
Using the Cairo 2D renderer in VTK. In progress.
-
New VTK view/representation class hierarchy.
-
New VTK-Qt view class hierarchy. Superceded by View Representation Refactoring.
-
Possible redesign of representations and views. Proposal.
-
Treemap view requirements. Completed.
-
Integrating the non-overlapping label placer into the view/representation scheme.
-
Discussion on how to integrate label and icon rendering.
-
Discussion on the various ways views could be linked.
-
Proposal of framework for views made up of other views.
-
Integrating vtkTreeMapView, vtkTreeRingView and vtkIcicleView. Completed.
-
Research about various treemap views and features.
Miscellaneous
|
How do I get support?
Send questions, bug reports or feature requests to vtkusers@vtk.org or vtk-developers@vtk.org.
- Someone from the Titan support and development team will get back to you with answers and we will log bugs or feature requests on the vtk bug tracker.
- Announcements are sent out to the vtkusers@vtk.org list so interested users should sign up for that list.
VisWeek 2008 Tutorial
The Titan team gave a half-day tutorial at IEEE Vis 2008. The tutorial contains an overview of the project and presents numerous examples in Python, C++, and Java of how to use the broad range of functionality within the toolkit.
Legacy Pages
These pages are outdated. They are kept for historical reasons.