If you had read the QL manuals or reference books, you would be aware SQL Server doesn't do multidimensional arrays.
I prefer BLOBs or VARCHAR(MAXCHAR) as an alternative. I enter the values in formatted string values delimiter separated.
But how to deal with multidimensional arrays. Well up to 3-5 you can use simple value for each "dimension". So the lowest dimension
would be comma delimit. The next set of values and each succeeding set for each 2nd dimension values would have the "set" of values separated
by a ":" or semicolon. For the third I use the ":" or colon to separate each 2D block of the 3rd dimension. This can all be easily reparsed
with a multidimensional parser for the application. Should be simple to build. (e.g.. Use string arrays to parse the higher dimensions
and numeric array for the lowest dimension). The VARCHAR(MAXCHAR) has an upper data bound of 2GB! If you have to save more
than 3 dimensions for most all applications, you don't understand how best to analyze the test data and may want to give it more
thought to optimize the operation as well as the design, so it doesn't get too complex. By the way I use Data Clusters built in the
data order and type as for SQL Server Data Table. This makes the design of the SQL Server interface easy as pie :).
