Sometimes it is best to skip writing a file at all. Your program
can send its data directly to the visualization or send its data to a database.
We know databases store all sorts of administrative information for businesses,
the government, and universities. Those same databases can store your scientific
data as well.
Databases optimize access to relational data.
- Metadata- You could store your dataset as
one huge chunk of information in the database and associate with that dataset information
about when you created the dataset or which program made it.p
- Tables of information - Maybe your data is,
itself, relational. For instance, finite element data relates points to neighbors
through edges and faces. Using a database to store this information could simplify
retrieval and manipulation.p
Databases store huge amounts of data. This is what they were made to do. You can distribute a dataset over linked SQL servers which will
look, to your program, like a single server.p
We have SQL Server running and could help you try it for your application.
