Graph tools
Help


Graph Tools

The graph tools are used to layout graphs according to different algorithms. They can only be used to layout normal graphs, and won't work with 3 dimensional graphs. They are layed out automaticly.

When laying out graphs please note that there is sometimes a problem with the view of the graph not updating properly. To get the view to update move any part of the graph.

The following describtion is derived from the JGraph manual, as the implementation of graphs is based on JGraph. For more indepth information please refer to your search engine of choice.

GEM

Radial Tree

Simple Circle

The circle layout arranges all the node into a circle, with constant spacing between each neighbour node.

The performance of this layout is proportional to the number of vertices in the circle. Although, circle layouts are not commonly used by themselves, it has been noted that some non-deterministic layouts (force-directed mainly) produce a better result if separated out by a circle layout first.

Annealing

Moen's Tree

The Moen Tree Layout is another layout in the tree-family, it makes some improvements over standard tree layouts. The Moen Tree takes cell shapes into account and concentrates on producing as compact a result as possible.

The Moen Tree also describes mechanisms to compute deltas of the layout, so the entire computation does not have to be performed on every layout. The Moen Tree manages to compact more tightly than the standard tree by storing sub-trees as polygons. In terms of performance the time to lay out using the layout is O(|V|), i.e. proportion to the number of vertices.

Sugiyama

Tree Layout

Spring Embedded

The Spring Layout is a force-directed layout algorithm designed to simulate a system of particles each with some mass. The vertices simulate mass points repelling each other and the edges simulate springs with attracting forces. The algorithm moves through a number of iteration trying to minimize the energy of this physical system.

This means a certain number of iterations are required to bring the system close to equilibrium, however, further iterations will perform very small changes and simply waste CPU time.

The performance of the Spring layout is O(|V|^2), i.e. proportional to the number of vertices squared. This time also needs to be multiplied by the number of iterations in the layout to get the full time worst-case. Generally, the spring is best applied to smaller graphs with a more regular structure.

A word of caution

Using the graph tools might cause exceptions to occur. Though these should not hamper the use of BRITNeY Suite.

Version 2, Thu 08 Jun 2006 11:41:08 [tveon] - created Tue 06 Jun 2006 16:10:42 [tveon]