| [eavp-dev] Plan for in-situ visualization in EAVP |
|
All,
I've been thinking about the role in-situ visualization can play in EAVP. I feel that we have our bases covered for pre- and post-processing through our ParaView and VisIt support and the geometry and mesh editors offer a good framework that we can expand
to meet the needs of individual projects. In-situ visualization is of great value to scientists, who benefit from the ability to get their results as soon as possible for the purposes of sanity checking long running simulations. ParaView and VisIt already
support in-situ visualization, so offering this capability through EAVP's connections to these programs should be both relatively simple and of interest to the customers already making use of them.
Some other projects in which we can leverage in-situ visualization include:
1) The sintering project should be able to leverage the geometry editor in displaying the object during the process. Current goals would be to set up a way to hook into the code to EAVP so that the geometry can be displayed as it is deformed and likely to
allow a texture to be set to a shape so that we can display scalar values (ie heat) for the shape. Other values, such as one that apply to the entire shape, might be placed in the Properties View or as graphs in subscenes. I'm unsure what values would be of
interest (or possible) to visualize, so guidance here would be appreciated. 2) ICEMAN. I've been told that there will be molecular simulations in ICEMAN, but I'm unsure of what actual characteristics are expected to change over time. I believe that this might simply require a molecule editor be implemented with the geometry editor,
with the molecule's shape being updated in real time but would need to be briefed about what the exact goal is for the project and where visualization could come into play for it. 3) Quantum computing is a novel area of visualization for EAVP. A brief search seems to turn up a program called jQuantum as a Java based visualizer for in-situ quantum computing. We should investigate this and see if it is suitable for use with quantum
research codes and if it can be integrated into EAVP. If not, then a custom visualization solution to show the computer's state as a quantum algorithm runs can be created instead
Robert |