2. The B2000++ Database

For reasons of computational efficiency, B2000++ programs, with the exception of the input processor, do not work with an MDL file, but instead with a so-called B2000++ database. Such a database is a directory containing sub-directories and files, it can be recognized by its suffix .b2m. A B2000++ database always contains the problem definition. If the b2000++ program has been run, the database contains the dof fields and the sampling/gradient fields as well.

The following table describes how the different programs access the B2000++ database:

Program Role
b2000++ Create a database from an MDL file (by calling b2ip++). Carry out the analysis and write dof fields and sampling/gradients fields to the database.
b2ip++ Create a database from an MDL file.
b2mass Write volume, mass, and center-of-gravity to the database.
b2mcbrowser Browses through MemCom datasets of a database, as well as individual datasets, and display them numerically.
b2rmdb Removes a B2000++ database.
baspl++ (B2000++ Pro) Read the mesh and the dof and gradient fields from the database.

The files and directories inside a B2000++ database (in this example, the database is called demo.b2m) can be listed using the find shell command:

$ find demo.b2m

The file log.txt contains log messages that were emitted during the analysis with B2000++. See the reference for the b2000++ command on how to customize log messages.

All mesh and solution data is stored within the subdirectory archives.mc, which represents the MemCom database. These files should not be manipulated by hand directly but rather by using one of the commands that are listed in the table above.

All programs that read from or write to a B2000++ database follow a set of conventions regarding the dataset names, dataset types, and dataset contents. These conventions are described in the section B2000++ Datasets of the programming manual.