From: [EMAIL PROTECTED]
> One fairly simple technique is to create a single table to contain only the
> key column(s) of the several partitioned tables; then use an insert and
> delete trigger on each of the main tables to maintain this table in tandem
> with them.
You still have to connect to one spot to do the insert, which brings in
locks and stuff.
Our cheat is the following.
insertQuery = connection.prepareStatement
("INSERT INTO "+ memberTable
+ " Values(?,"+
"(select ((((COALESCE(max(M.mr_id),0))/100)+1)*100) + "+
"max(D.db_id) from "+ memberTable +" as M, "+
databaseTable+ " as D)"+
",?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?)");
Where the databaseTable has a single unique ID for each database (1-99) and
then the records get replicated around as needed.
You could do the same with a unique partition table for each partition.
-HJC
-
::: When replying to the list, please use 'Reply-All' and make sure
::: a copy goes to the list ([EMAIL PROTECTED]).
*** To unsubscribe, send 'unsubscribe' to [EMAIL PROTECTED]
*** For more information, check http://www.db2eug.uni.cc
