Is it necessary to read all the data back into LabVIEW first? The reason I'm asking is that it is much more efficient to just use SQL statements to get the data you need and write that directly to the second database. I did a quick Google search on it and came up with lots of responses on what the correct syntax is for setting that up. Using an SQL command to copy the data back and forth is analogous to copying data from one file to another using the Copy file function rather than reading the entire contents of the file, converting the file string to the data arrays, converting back to a string, and then writing it back to another file.
Chapter 5 of the Database Connectivity Toolset User Manual describes how to execute SQL queries.
If you do need to actually
see the data in LabVIEW first, then the only way to get it back into the other database is to convert those variants to LV datatypes and then cluster those values again and write them back out to the second database.