The Fiber Bundle Data Model

Data within HydroVish are primarily organised as "fiber bundles", which is an originally mathematical concept from topology. The idea to use fiber bundle stems from a visionary scientific article in 1998 advocating to use this mathematical scheme as a common concept for all kinds of data, rather than treating each type of data independently and on its own. This idea has been implemented in HydroVish and developed further for practical usage. To the most part this data organisation scheme is invisible to the user, "things just work", but on certain aspects it shines through, so it is helpful for the user to at least have a clue to know what it is about and how it works.

Basically a fiber bundle is a pair of a "base space" - think of it as some geometric entity - with a "fiber" attached to each point. This "fiber" is any kind of attribute information per point, such as color, temperature, brightness or anything more.

In the HydroVish fiber bundle data model geometric entities are organized into Grid objects with attribute information living on Fields. A general data set consists of many Grids stored in a Bundle. There is no limit on how many Grids may live on a Bundle, and there is no limit on how many Fields may live on each Grid.

Within a Bundle a Grid is selected via a name that the user is free to specify according to their needs when creating the data set. For instance it is useful to use a name that refers to the origin of the data, such as the name of an project, or some geographical region. In many cases, a Bundle may contain only a single Grid, though the infrastructure supports many, which e.g. allows easy switching and comparisons between projects, regions, etc.

Within a Grid a Field is selected via a name that describes the meaning of this information. While technically arbitrary, various applications introduce certain conventions on how to call which information.

Visualization and data processing modules may operate on either Grid or Field objects, depending on how much information they require. When reading a data file, all data will be provided as a Bundle:

This Bundle object provides basic information such as how many Grid objects are contained here. For most subsequent operatoins, a specific Grid needs to be extracted, and possibly Field from this Grid. This is done via the Fiber context menu in the output section or via the [F] shortcut button:

This initial setup constitutes a first basic visualization network upon which more application specific setups can be build.

For instance, it is recommended to display a bounding on Grid objects as a first visual hint about the spatial extent of a newly loaded dataset. Also, the Display/HeightDots module may provide good first views on Grid objects.

Starting from these initial steps, more advanced visualization networks are ready to be constructed, such as in the Editing Tutorial.

File I/O


For an overview of all VISH help topics see the Table of Contents.