Hi Christian,
Derby does not support partitioned tables. I think you will need to
write your own logic to divide the attachments among many buckets--the
buckets could either be a growing number of tables inside the database
or files outside the database. Regardless of how you implement the
buckets, table functions may be able to hide some of the complexity when
you SELECT the attachments. You may be able to use database procedures
to hide some of the complexity of INSERTing into the buckets.
Hope this helps,
-Rick
Christian Gruber wrote:
Hi,
the maximum size of a table is the maximum file size of the underlying
operating system (e.g.: 2 GB).
For data with a size of 10 MB I would reach the limit with about 200
entries.
The data once in the table needs only be read (attachments of
messages). Perhaps the new table function can be used to show the
content of the filesystem as a table, when the attachments are stored
outside of the database.
Is there a way to distribute a table over more than one file or
another solution?
Any solution is appreciated!
Thanks
Christian
