This page describes additions to VisIt for visualizing Embedded Boundary(EB) CHOMBO data.
The following features have been added to VisIt to support EB CHOMBO.
- An EB CHOMBO database reader that supersedes the existing CHOMBO database reader.
- An glyphing operator that can glyph edges and nodes of the graph
- A channel generating operator that generates geometry for cut cells, filter covered cells, and generates shoreline
- A flood fill operator that takes data from a low resolution multilevel simulation dataset and injects it into a high resolution single level dataset.
These additions are described in depth in the following sections.
The new EB CHOMBO database reader supersedes the existing CHOMBO database reader and is EB aware. EB datasets contain the standard rectilinear mutilevel AMR hierarchy and an additional graph that describes the geometry of cut cells. CHOMBO simulation will need to write the data in the new format to take advantage of the new database reader.
In terms of VisIt's internal organization a database can provide a number of meshes, and variables associated with particular meshes. The mechanism used internally to identify the mesh that a given variable comes from is through a heuristic encoded in the variable name. The rectilinear mesh and it's data are accessed via the Mesh identifier while the graph and it's associated data are accessed via the Eb Graph identifier. The EB CHOMBO database can be used to access either of these as needed at the time a plot is created. This is demonstrated in figures 1 and 2.
The addition of the EB Graph mesh brings some complexity to selecting variables. In the old CHOMBO format variables could be defined on cell, face, or node centers, and VisIt assumed cell centering. In the new EB CHOMBO format data defined on regular and covered cells can be cell, face or node centered. Additionally, the data from irregular cells can be node or cell centered on the EB Graph. Keep in mind that the terms cell centered and node centered are given with respect to a particular mesh. For example, a cell center variable in the rectilinear data is node centered on the EB Graph data.
Two important plots that can be made after loading an EB CHOMBO file are Mesh and Pseudocolor plots. The Mesh plot allows one to visualize the edges of the Eb Graph. The Pseudocolor plot allows one to color graph edges by any of the variables. These two plots provide basic functionality that can be augmented by further applying the new operators described below.